Request Headers
Authorization token (required).
Request Body
The retry token provided by a previous incomplete or failed profile computation.
If true, vector completion tasks use cached votes from the global ObjectiveAI votes cache when available. Has lower priority than retry_token, higher priority than from_rng.
If true, any remaining votes from vector completion tasks are generated via RNG. Has lower priority than retry_token or from_cache.
The maximum number of retries to attempt when a function execution fails during profile computation.
The number of function executions to perform per dataset item. Generally speaking, higher N values increase the quality of the computed profile.
The dataset of input and target output pairs to use for computing the profile.
Items
A Function input and its corresponding target output.
Properties
A value provided as input to a function.
Variants
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
Values
Items
The target output for a given function input.
Variants
A scalar target output. The desired output is this exact scalar.
Properties
A vector target output. The desired output is this exact vector.
Properties
Items
A vector winner target output. The desired output is a vector where the highest value is at the specified index.
Properties
The Ensemble to use for this completion. May be a unique ID or an inline definition.
Variants
The unique identifier for the Ensemble.
An ensemble of LLMs.
Properties
The list of LLMs that make up the ensemble.
Items
An LLM to be used within an Ensemble, including optional fallbacks and count.
Properties
The full ID of the LLM to use.
For Vector Completions only, specifies the LLM's voting output mode. For "instruction", the assistant is instructed to output a key. For "json_schema", the assistant is constrained to output a valid key using a JSON schema. For "tool_call", the assistant is instructed to output a tool call to select the key.
Variants
For Vector Completions only, whether to use synthetic reasoning prior to voting. Works for any LLM, even those that do not have native reasoning capabilities.
For Vector Completions only, whether to use logprobs to make the vote probabilistic. This means that the LLM can vote for multiple keys based on their logprobabilities. Allows LLMs to express native uncertainty when voting.
A list of messages exchanged in a chat conversation. These will be prepended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
A list of messages exchanged in a chat conversation. These will be appended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Values
An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
This setting aims to control the presence of tokens in the output. It tries to encourage the model to use tokens that are less present in the input, proportional to their presence in the input. Token presence scales with the number of occurrences. Negative values will encourage more diverse token usage.
The assistant will stop when any of the provided strings are generated.
Variants
Generation will stop when this string is generated.
Generation will stop when any of these strings are generated.
Items
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Options for selecting the upstream provider of this model.
Properties
Whether to allow fallback providers if the preferred provider is unavailable.
Whether to require that the provider supports all specified parameters.
An ordered list of provider names to use when selecting a provider for this model.
Items
A list of provider names to restrict selection to when selecting a provider for this model.
Items
A list of provider names to ignore when selecting a provider for this model.
Items
Specifies the quantizations to allow when selecting providers for this model.
Items
Options for controlling reasoning behavior of the model.
Properties
Enables or disables reasoning for supported models.
The maximum number of tokens to use for reasoning in a response.
Constrains effort on reasoning for supported reasoning models. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
Variants
Controls the verbosity of the reasoning summary for supported reasoning models.
Variants
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.
Variants
A count greater than one effectively means that there are multiple instances of this LLM in an ensemble.
A list of fallback LLMs to use if the primary LLM fails.
Items
An LLM to be used within an Ensemble or standalone with Chat Completions.
Properties
The full ID of the LLM to use.
For Vector Completions only, specifies the LLM's voting output mode. For "instruction", the assistant is instructed to output a key. For "json_schema", the assistant is constrained to output a valid key using a JSON schema. For "tool_call", the assistant is instructed to output a tool call to select the key.
Variants
For Vector Completions only, whether to use synthetic reasoning prior to voting. Works for any LLM, even those that do not have native reasoning capabilities.
For Vector Completions only, whether to use logprobs to make the vote probabilistic. This means that the LLM can vote for multiple keys based on their logprobabilities. Allows LLMs to express native uncertainty when voting.
A list of messages exchanged in a chat conversation. These will be prepended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
A list of messages exchanged in a chat conversation. These will be appended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Values
An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
This setting aims to control the presence of tokens in the output. It tries to encourage the model to use tokens that are less present in the input, proportional to their presence in the input. Token presence scales with the number of occurrences. Negative values will encourage more diverse token usage.
The assistant will stop when any of the provided strings are generated.
Variants
Generation will stop when this string is generated.
Generation will stop when any of these strings are generated.
Items
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Options for selecting the upstream provider of this model.
Properties
Whether to allow fallback providers if the preferred provider is unavailable.
Whether to require that the provider supports all specified parameters.
An ordered list of provider names to use when selecting a provider for this model.
Items
A list of provider names to restrict selection to when selecting a provider for this model.
Items
A list of provider names to ignore when selecting a provider for this model.
Items
Specifies the quantizations to allow when selecting providers for this model.
Items
Options for controlling reasoning behavior of the model.
Properties
Enables or disables reasoning for supported models.
The maximum number of tokens to use for reasoning in a response.
Constrains effort on reasoning for supported reasoning models. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
Variants
Controls the verbosity of the reasoning summary for supported reasoning models.
Variants
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.
Variants
Options for selecting the upstream provider of this completion.
Properties
Specifies whether to allow providers which collect data.
Variants
Whether to enforce Zero Data Retention (ZDR) policies when selecting providers.
Specifies the sorting strategy for provider selection.
Variants
Properties
Maximum price for prompt tokens.
Maximum price for completion tokens.
Maximum price for image generation.
Maximum price for audio generation.
Maximum price per request.
Preferred minimum throughput for the provider.
Preferred maximum latency for the provider.
Minimum throughput for the provider.
Maximum latency for the provider.
If specified, upstream systems will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result.
The maximum total time in milliseconds to spend on retries when a transient error occurs.
The maximum time in milliseconds to wait for the first chunk of a streaming response.
The maximum time in milliseconds to wait between subsequent chunks of a streaming response.
A function defined inline.
Variants
A scalar function defined inline.
Properties
An expression or list of expressions which evaluate to a 2D array of Inputs. Each sub-array will be fed into Tasks which specify an index of this input map.
Variants
An expression which evaluates to a 2D array of Inputs.
Properties
A JMESPath expression.
A list of expressions which each evaluate to a 1D array of Inputs.
Items
An expression which evaluates to a 1D array of Inputs.
Properties
A JMESPath expression.
The list of tasks to be executed as part of the function. Each will first be compiled using the parent function's input.
Items
A scalar function task expression.
Properties
The owner of the GitHub repository containing the function.
The name of the GitHub repository containing the function.
The commit SHA of the GitHub repository containing the function.
An expression which evaluates to a boolean indicating whether to skip this task.
Properties
A JMESPath expression.
If present, indicates that this task should be ran once for each entry in the specified input map (input map is a 2D array indexed by this value).
A value provided as input to a function.
Variants
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
Values
Items
An expression which evaluates to an input value.
Properties
A JMESPath expression.
A vector function task expression.
Properties
The owner of the GitHub repository containing the function.
The name of the GitHub repository containing the function.
The commit SHA of the GitHub repository containing the function.
An expression which evaluates to a boolean indicating whether to skip this task.
Properties
A JMESPath expression.
If present, indicates that this task should be ran once for each entry in the specified input map (input map is a 2D array indexed by this value).
A value provided as input to a function.
Variants
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
Values
Items
An expression which evaluates to an input value.
Properties
A JMESPath expression.
A vector completion task expression.
Properties
An expression which evaluates to a boolean indicating whether to skip this task.
Properties
A JMESPath expression.
If present, indicates that this task should be ran once for each entry in the specified input map (input map is a 2D array indexed by this value).
A list of messages exchanged in a chat conversation.
Variants
A list of messages exchanged in a chat conversation.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
Properties
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An expression which evaluates to a simple content part.
Properties
A JMESPath expression.
An expression which evaluates to simple content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
Properties
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An expression which evaluates to a simple content part.
Properties
A JMESPath expression.
An expression which evaluates to simple content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Variants
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An expression which evaluates to the image URL definition.
Properties
A JMESPath expression.
An audio rich content part.
Properties
The audio data and its format.
Variants
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
An expression which evaluates to the audio definition.
Properties
A JMESPath expression.
A video rich content part.
Properties
Variants
The video URL definition.
Variants
Properties
URL of the video.
An expression which evaluates to the video URL definition.
Properties
A JMESPath expression.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Variants
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to the file definition.
Properties
A JMESPath expression.
An expression which evaluates to a rich content part.
Properties
A JMESPath expression.
An expression which evaluates to rich content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Variants
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An expression which evaluates to the image URL definition.
Properties
A JMESPath expression.
An audio rich content part.
Properties
The audio data and its format.
Variants
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
An expression which evaluates to the audio definition.
Properties
A JMESPath expression.
A video rich content part.
Properties
Variants
The video URL definition.
Variants
Properties
URL of the video.
An expression which evaluates to the video URL definition.
Properties
A JMESPath expression.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Variants
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to the file definition.
Properties
A JMESPath expression.
An expression which evaluates to a rich content part.
Properties
A JMESPath expression.
An expression which evaluates to rich content.
Properties
A JMESPath expression.
The ID of the tool call that this message is responding to.
Variants
The ID of the tool call that this message is responding to.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Variants
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An expression which evaluates to the image URL definition.
Properties
A JMESPath expression.
An audio rich content part.
Properties
The audio data and its format.
Variants
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
An expression which evaluates to the audio definition.
Properties
A JMESPath expression.
A video rich content part.
Properties
Variants
The video URL definition.
Variants
Properties
URL of the video.
An expression which evaluates to the video URL definition.
Properties
A JMESPath expression.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Variants
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to the file definition.
Properties
A JMESPath expression.
An expression which evaluates to a rich content part.
Properties
A JMESPath expression.
An expression which evaluates to rich content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The refusal message by the assistant.
Variants
The refusal message by the assistant.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Tool calls made by the assistant.
Variants
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
Variants
The unique identifier for the tool call.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The name and arguments of the function called.
Properties
The name of the function called.
Variants
The name of the function called.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The arguments passed to the function.
Variants
The arguments passed to the function.
An expression which evaluates to a string.
Properties
A JMESPath expression.
An expression which evaluates to a tool call.
Properties
A JMESPath expression.
An expression which evaluates to an array of tool calls.
Properties
A JMESPath expression.
The reasoning provided by the assistant.
Variants
The reasoning provided by the assistant.
An expression which evaluates to a string.
Properties
A JMESPath expression.
An expression which evaluates to a message.
Properties
A JMESPath expression.
An expression which evaluates to an array of messages.
Properties
A JMESPath expression.
A list of tools that the assistant can call. These are readonly and will only be useful for explaining prior tool calls or otherwise influencing behavior.
Variants
A list of tools that the assistant can call.
Items
A function tool that the assistant can call.
Properties
The definition of a function tool.
Properties
The name of the function.
Variants
The name of the function.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The description of the function.
Variants
The description of the function.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The JSON schema defining the parameters of the function.
Variants
Values
Null
Boolean
Number
String
Array (values may be expressions)
Items
A JSON value.
Object (values may be expressions)
Values
A JSON value.
An expression which evaluates to a JSON value.
Properties
A JMESPath expression.
An expression which evaluates to a JSON schema object.
Properties
A JMESPath expression.
Whether to enforce strict adherence to the parameter schema.
Variants
Whether to enforce strict adherence to the parameter schema.
An expression which evaluates to a boolean.
Properties
A JMESPath expression.
An expression which evaluates to a tool.
Properties
A JMESPath expression.
An expression which evaluates to an array of tools.
Properties
A JMESPath expression.
A list of possible assistant responses which the LLMs in the Ensemble will vote on. The output scores will be of the same length, each corresponding to one response. The winner is the response with the highest score.
Variants
A list of possible assistant responses which the LLMs in the Ensemble will vote on. The output scores will be of the same length, each corresponding to one response. The winner is the response with the highest score.
Items
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to a possible assistant response.
Properties
A JMESPath expression.
An expression which evaluates to an array of possible assistant responses.
Properties
A JMESPath expression.
An expression which evaluates to a single number. This is the output of the scalar function. Will be provided with the outputs of all tasks.
Properties
A JMESPath expression.
A vector function defined inline.
Properties
An expression or list of expressions which evaluate to a 2D array of Inputs. Each sub-array will be fed into Tasks which specify an index of this input map.
Variants
An expression which evaluates to a 2D array of Inputs.
Properties
A JMESPath expression.
A list of expressions which each evaluate to a 1D array of Inputs.
Items
An expression which evaluates to a 1D array of Inputs.
Properties
A JMESPath expression.
The list of tasks to be executed as part of the function. Each will first be compiled using the parent function's input.
Items
A scalar function task expression.
Properties
The owner of the GitHub repository containing the function.
The name of the GitHub repository containing the function.
The commit SHA of the GitHub repository containing the function.
An expression which evaluates to a boolean indicating whether to skip this task.
Properties
A JMESPath expression.
If present, indicates that this task should be ran once for each entry in the specified input map (input map is a 2D array indexed by this value).
A value provided as input to a function.
Variants
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
Values
Items
An expression which evaluates to an input value.
Properties
A JMESPath expression.
A vector function task expression.
Properties
The owner of the GitHub repository containing the function.
The name of the GitHub repository containing the function.
The commit SHA of the GitHub repository containing the function.
An expression which evaluates to a boolean indicating whether to skip this task.
Properties
A JMESPath expression.
If present, indicates that this task should be ran once for each entry in the specified input map (input map is a 2D array indexed by this value).
A value provided as input to a function.
Variants
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
Values
Items
An expression which evaluates to an input value.
Properties
A JMESPath expression.
A vector completion task expression.
Properties
An expression which evaluates to a boolean indicating whether to skip this task.
Properties
A JMESPath expression.
If present, indicates that this task should be ran once for each entry in the specified input map (input map is a 2D array indexed by this value).
A list of messages exchanged in a chat conversation.
Variants
A list of messages exchanged in a chat conversation.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
Properties
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An expression which evaluates to a simple content part.
Properties
A JMESPath expression.
An expression which evaluates to simple content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
Properties
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An expression which evaluates to a simple content part.
Properties
A JMESPath expression.
An expression which evaluates to simple content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Variants
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An expression which evaluates to the image URL definition.
Properties
A JMESPath expression.
An audio rich content part.
Properties
The audio data and its format.
Variants
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
An expression which evaluates to the audio definition.
Properties
A JMESPath expression.
A video rich content part.
Properties
Variants
The video URL definition.
Variants
Properties
URL of the video.
An expression which evaluates to the video URL definition.
Properties
A JMESPath expression.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Variants
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to the file definition.
Properties
A JMESPath expression.
An expression which evaluates to a rich content part.
Properties
A JMESPath expression.
An expression which evaluates to rich content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Variants
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An expression which evaluates to the image URL definition.
Properties
A JMESPath expression.
An audio rich content part.
Properties
The audio data and its format.
Variants
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
An expression which evaluates to the audio definition.
Properties
A JMESPath expression.
A video rich content part.
Properties
Variants
The video URL definition.
Variants
Properties
URL of the video.
An expression which evaluates to the video URL definition.
Properties
A JMESPath expression.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Variants
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to the file definition.
Properties
A JMESPath expression.
An expression which evaluates to a rich content part.
Properties
A JMESPath expression.
An expression which evaluates to rich content.
Properties
A JMESPath expression.
The ID of the tool call that this message is responding to.
Variants
The ID of the tool call that this message is responding to.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
Variants
The text content.
An expression which evaluates to the text content.
Properties
A JMESPath expression.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Variants
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An expression which evaluates to the image URL definition.
Properties
A JMESPath expression.
An audio rich content part.
Properties
The audio data and its format.
Variants
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
An expression which evaluates to the audio definition.
Properties
A JMESPath expression.
A video rich content part.
Properties
Variants
The video URL definition.
Variants
Properties
URL of the video.
An expression which evaluates to the video URL definition.
Properties
A JMESPath expression.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Variants
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to the file definition.
Properties
A JMESPath expression.
An expression which evaluates to a rich content part.
Properties
A JMESPath expression.
An expression which evaluates to rich content.
Properties
A JMESPath expression.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Variants
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The refusal message by the assistant.
Variants
The refusal message by the assistant.
An expression which evaluates to a string.
Properties
A JMESPath expression.
Tool calls made by the assistant.
Variants
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
Variants
The unique identifier for the tool call.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The name and arguments of the function called.
Properties
The name of the function called.
Variants
The name of the function called.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The arguments passed to the function.
Variants
The arguments passed to the function.
An expression which evaluates to a string.
Properties
A JMESPath expression.
An expression which evaluates to a tool call.
Properties
A JMESPath expression.
An expression which evaluates to an array of tool calls.
Properties
A JMESPath expression.
The reasoning provided by the assistant.
Variants
The reasoning provided by the assistant.
An expression which evaluates to a string.
Properties
A JMESPath expression.
An expression which evaluates to a message.
Properties
A JMESPath expression.
An expression which evaluates to an array of messages.
Properties
A JMESPath expression.
A list of tools that the assistant can call. These are readonly and will only be useful for explaining prior tool calls or otherwise influencing behavior.
Variants
A list of tools that the assistant can call.
Items
A function tool that the assistant can call.
Properties
The definition of a function tool.
Properties
The name of the function.
Variants
The name of the function.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The description of the function.
Variants
The description of the function.
An expression which evaluates to a string.
Properties
A JMESPath expression.
The JSON schema defining the parameters of the function.
Variants
Values
Null
Boolean
Number
String
Array (values may be expressions)
Items
A JSON value.
Object (values may be expressions)
Values
A JSON value.
An expression which evaluates to a JSON value.
Properties
A JMESPath expression.
An expression which evaluates to a JSON schema object.
Properties
A JMESPath expression.
Whether to enforce strict adherence to the parameter schema.
Variants
Whether to enforce strict adherence to the parameter schema.
An expression which evaluates to a boolean.
Properties
A JMESPath expression.
An expression which evaluates to a tool.
Properties
A JMESPath expression.
An expression which evaluates to an array of tools.
Properties
A JMESPath expression.
A list of possible assistant responses which the LLMs in the Ensemble will vote on. The output scores will be of the same length, each corresponding to one response. The winner is the response with the highest score.
Variants
A list of possible assistant responses which the LLMs in the Ensemble will vote on. The output scores will be of the same length, each corresponding to one response. The winner is the response with the highest score.
Items
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An expression which evaluates to a possible assistant response.
Properties
A JMESPath expression.
An expression which evaluates to an array of possible assistant responses.
Properties
A JMESPath expression.
An expression which evaluates to an array of numbers. This is the output of the vector function. Will be provided with the outputs of all tasks.
Properties
A JMESPath expression.
An expression transforming input into an array of inputs. When the Function is executed with any input from the array, the output_length should be 1. Only required if the request uses a strategy that needs input splitting (e.g., swiss_system).
Properties
A JMESPath expression.
An expression transforming an array of inputs (computed by input_split) into a single Input object for the Function. Only required if the request uses a strategy that needs input splitting (e.g., swiss_system).
Properties
A JMESPath expression.
Whether to stream the response as a series of chunks.
Response Body
The unique identifier of the function profile computation.
The function executions performed as part of computing the profile.
Items
A function execution ran during profile computation.
Properties
The unique identifier of the function execution.
The tasks executed as part of the function execution.
Items
A function execution task.
Properties
The unique identifier of the function execution.
The tasks executed as part of the function execution.
When true, indicates that one or more tasks encountered errors during execution.
A reasoning summary generation.
Properties
The unique identifier of the chat completion.
The unique identifier of the upstream chat completion.
The list of choices in this chat completion.
Items
A choice in a unary chat completion response.
Properties
A message generated by the assistant.
Properties
The content of the message.
The refusal message, if any.
The role of the message author.
Variants
The tool calls made by the assistant, if any.
Items
A function tool call made by the assistant.
Properties
The unique identifier of the function tool.
Properties
The name of the function.
The arguments passed to the function.
The reasoning provided by the assistant, if any.
The images generated by the assistant, if any.
Items
Properties
Properties
The Base64 URL of the generated image.
The reason why the assistant ceased to generate further tokens.
Variants
The index of the choice in the list of choices.
The log probabilities of the tokens generated by the model.
Properties
The log probabilities of the tokens in the content.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The log probabilities of the tokens in the refusal.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The Unix timestamp (in seconds) when the chat completion was created.
The unique identifier of the Ensemble LLM used for this chat completion.
The upstream model used for this chat completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The cost multiplier applied to upstream costs for computing ObjectiveAI costs.
Whether the completion used a BYOK (Bring Your Own Key) API Key.
The provider used for this chat completion.
When non-null, indicates that an error occurred during the chat completion.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
The output of the function execution.
Variants
The scalar output of the function execution.
The vector output of the function execution.
Items
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
When non-null, indicates that an error occurred during the function execution.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
A token which may be used to retry the function execution.
The UNIX timestamp (in seconds) when the function execution chunk was created.
The unique identifier of the function being executed.
The unique identifier of the profile being used.
The object type.
Variants
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The index of the task in the sequence of tasks.
The index of the task amongst all mapped and non-skipped compiled tasks. Used internally.
The path of this task which may be used to navigate which nested task this is amongst the root functions tasks and sub-tasks.
Items
The Swiss system round number (1-indexed).
The index of this task within its Swiss system pool.
A vector completion task.
Properties
The unique identifier of the vector completion.
The list of chat completions created for this vector completion.
Items
A chat completion generated in the pursuit of a vector completion.
Properties
The unique identifier of the chat completion.
The unique identifier of the upstream chat completion.
The list of choices in this chat completion.
Items
A choice in a unary chat completion response.
Properties
A message generated by the assistant.
Properties
The content of the message.
The refusal message, if any.
The role of the message author.
Variants
The tool calls made by the assistant, if any.
Items
A function tool call made by the assistant.
Properties
The unique identifier of the function tool.
Properties
The name of the function.
The arguments passed to the function.
The reasoning provided by the assistant, if any.
The images generated by the assistant, if any.
Items
Properties
Properties
The Base64 URL of the generated image.
The reason why the assistant ceased to generate further tokens.
Variants
The index of the choice in the list of choices.
The log probabilities of the tokens generated by the model.
Properties
The log probabilities of the tokens in the content.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The log probabilities of the tokens in the refusal.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The Unix timestamp (in seconds) when the chat completion was created.
The unique identifier of the Ensemble LLM used for this chat completion.
The upstream model used for this chat completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The cost multiplier applied to upstream costs for computing ObjectiveAI costs.
Whether the completion used a BYOK (Bring Your Own Key) API Key.
The provider used for this chat completion.
The index of the completion amongst all chat completions.
An error encountered during the generation of this chat completion.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
The list of votes for responses in the request from the Ensemble LLMs within the provided Ensemble.
Items
A vote from an Ensemble LLM within a Vector Completion.
Properties
The unique identifier of the Ensemble LLM which generated this vote.
The index of the Ensemble LLM in the Ensemble.
The flat index of the Ensemble LLM in the expanded Ensemble, accounting for counts.
The vote generated by this Ensemble LLM. It is of the same length of the number of responses provided in the request. If the Ensemble LLM used logprobs, may be a probability distribution; otherwise, one of the responses will have a value of 1 and the rest 0.
Items
The weight assigned to this vote.
Whether this vote came from a previous Vector Completion which was retried. from_cache will also be true.
Whether this vote came from the global ObjectiveAI votes cache.
Whether this vote was generated via RNG.
The scores for each response in the request, aggregated from the votes of the Ensemble LLMs.
Items
The weights assigned to each response in the request, aggregated from the votes of the Ensemble LLMs.
Items
The Unix timestamp (in seconds) when the vector completion was created.
The unique identifier of the Ensemble used for this vector completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The index of the task in the sequence of tasks.
The index of the task amongst all mapped and non-skipped compiled tasks. Used internally.
The path of this task which may be used to navigate which nested task this is amongst the root functions tasks and sub-tasks.
Items
When non-null, indicates that an error occurred during the vector completion task.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
When true, indicates that one or more tasks encountered errors during execution.
A reasoning summary generation.
Properties
The unique identifier of the chat completion.
The unique identifier of the upstream chat completion.
The list of choices in this chat completion.
Items
A choice in a unary chat completion response.
Properties
A message generated by the assistant.
Properties
The content of the message.
The refusal message, if any.
The role of the message author.
Variants
The tool calls made by the assistant, if any.
Items
A function tool call made by the assistant.
Properties
The unique identifier of the function tool.
Properties
The name of the function.
The arguments passed to the function.
The reasoning provided by the assistant, if any.
The images generated by the assistant, if any.
Items
Properties
Properties
The Base64 URL of the generated image.
The reason why the assistant ceased to generate further tokens.
Variants
The index of the choice in the list of choices.
The log probabilities of the tokens generated by the model.
Properties
The log probabilities of the tokens in the content.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The log probabilities of the tokens in the refusal.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The Unix timestamp (in seconds) when the chat completion was created.
The unique identifier of the Ensemble LLM used for this chat completion.
The upstream model used for this chat completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The cost multiplier applied to upstream costs for computing ObjectiveAI costs.
Whether the completion used a BYOK (Bring Your Own Key) API Key.
The provider used for this chat completion.
When non-null, indicates that an error occurred during the chat completion.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
The output of the function execution.
Variants
The scalar output of the function execution.
The vector output of the function execution.
Items
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
When non-null, indicates that an error occurred during the function execution.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
A token which may be used to retry the function execution.
The UNIX timestamp (in seconds) when the function execution chunk was created.
The unique identifier of the function being executed.
The unique identifier of the profile being used.
The object type.
Variants
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The index of the function execution in the list of executions.
The index of the dataset item this function execution corresponds to.
The N index for this function execution. There will be N function executions, and N comes from the request parameters.
The retry index for this function execution. There may be multiple retries for a given dataset item and N index.
When true, indicates that one or more function executions encountered errors during profile computation.
A function profile defined inline.
Properties
The list of task profiles.
Items
The identifiers for a function profile hosted in a GitHub repository.
Properties
The owner of the GitHub repository containing the profile.
The name of the GitHub repository containing the profile.
The commit SHA of the GitHub repository containing the profile.
A function profile defined inline.
Properties
The list of task profiles.
Items
A vector completion profile for a vector completion task containing an Ensemble and array of weights.
Properties
The Ensemble to use for this completion. May be a unique ID or an inline definition.
Variants
The unique identifier for the Ensemble.
An ensemble of LLMs.
Properties
The list of LLMs that make up the ensemble.
Items
An LLM to be used within an Ensemble, including optional fallbacks and count.
Properties
The full ID of the LLM to use.
For Vector Completions only, specifies the LLM's voting output mode. For "instruction", the assistant is instructed to output a key. For "json_schema", the assistant is constrained to output a valid key using a JSON schema. For "tool_call", the assistant is instructed to output a tool call to select the key.
Variants
For Vector Completions only, whether to use synthetic reasoning prior to voting. Works for any LLM, even those that do not have native reasoning capabilities.
For Vector Completions only, whether to use logprobs to make the vote probabilistic. This means that the LLM can vote for multiple keys based on their logprobabilities. Allows LLMs to express native uncertainty when voting.
A list of messages exchanged in a chat conversation. These will be prepended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
A list of messages exchanged in a chat conversation. These will be appended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Values
An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
This setting aims to control the presence of tokens in the output. It tries to encourage the model to use tokens that are less present in the input, proportional to their presence in the input. Token presence scales with the number of occurrences. Negative values will encourage more diverse token usage.
The assistant will stop when any of the provided strings are generated.
Variants
Generation will stop when this string is generated.
Generation will stop when any of these strings are generated.
Items
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Options for selecting the upstream provider of this model.
Properties
Whether to allow fallback providers if the preferred provider is unavailable.
Whether to require that the provider supports all specified parameters.
An ordered list of provider names to use when selecting a provider for this model.
Items
A list of provider names to restrict selection to when selecting a provider for this model.
Items
A list of provider names to ignore when selecting a provider for this model.
Items
Specifies the quantizations to allow when selecting providers for this model.
Items
Options for controlling reasoning behavior of the model.
Properties
Enables or disables reasoning for supported models.
The maximum number of tokens to use for reasoning in a response.
Constrains effort on reasoning for supported reasoning models. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
Variants
Controls the verbosity of the reasoning summary for supported reasoning models.
Variants
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.
Variants
A count greater than one effectively means that there are multiple instances of this LLM in an ensemble.
A list of fallback LLMs to use if the primary LLM fails.
Items
An LLM to be used within an Ensemble or standalone with Chat Completions.
Properties
The full ID of the LLM to use.
For Vector Completions only, specifies the LLM's voting output mode. For "instruction", the assistant is instructed to output a key. For "json_schema", the assistant is constrained to output a valid key using a JSON schema. For "tool_call", the assistant is instructed to output a tool call to select the key.
Variants
For Vector Completions only, whether to use synthetic reasoning prior to voting. Works for any LLM, even those that do not have native reasoning capabilities.
For Vector Completions only, whether to use logprobs to make the vote probabilistic. This means that the LLM can vote for multiple keys based on their logprobabilities. Allows LLMs to express native uncertainty when voting.
A list of messages exchanged in a chat conversation. These will be prepended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
A list of messages exchanged in a chat conversation. These will be appended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Values
An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
This setting aims to control the presence of tokens in the output. It tries to encourage the model to use tokens that are less present in the input, proportional to their presence in the input. Token presence scales with the number of occurrences. Negative values will encourage more diverse token usage.
The assistant will stop when any of the provided strings are generated.
Variants
Generation will stop when this string is generated.
Generation will stop when any of these strings are generated.
Items
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Options for selecting the upstream provider of this model.
Properties
Whether to allow fallback providers if the preferred provider is unavailable.
Whether to require that the provider supports all specified parameters.
An ordered list of provider names to use when selecting a provider for this model.
Items
A list of provider names to restrict selection to when selecting a provider for this model.
Items
A list of provider names to ignore when selecting a provider for this model.
Items
Specifies the quantizations to allow when selecting providers for this model.
Items
Options for controlling reasoning behavior of the model.
Properties
Enables or disables reasoning for supported models.
The maximum number of tokens to use for reasoning in a response.
Constrains effort on reasoning for supported reasoning models. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
Variants
Controls the verbosity of the reasoning summary for supported reasoning models.
Variants
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.
Variants
The profile to use for the completion. Must be of the same length as the Ensemble's llms field, ignoring count.
Items
Statistics about the fitting process used to compute the weights for the profile.
Properties
The final sum loss achieved during weights fitting.
The total number of function executions used during weights fitting.
The number of fitting starts attempted. Each start begins with a randomized weight vector.
The number of fitting rounds performed across all starts.
The number of errors which occurred while computing outputs during fitting.
A token which may be used to retry the function profile computation.
The UNIX timestamp (in seconds) when the function profile computation was created.
The unique identifier of the function for which the profile is being computed.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
Response Body (Streaming)
The unique identifier of the function profile computation chunk.
The function executions performed as part of computing the profile.
Items
A chunk of a function execution ran during profile computation.
Properties
The unique identifier of the function execution.
The tasks executed as part of the function execution.
Items
A chunk of a function execution task.
Properties
The unique identifier of the function execution.
The tasks executed as part of the function execution.
When true, indicates that one or more tasks encountered errors during execution.
A chunk of a reasoning summary generation.
Properties
The unique identifier of the chat completion.
The unique identifier of the upstream chat completion.
The list of choices in this chunk.
Items
A choice in a streaming chat completion response.
Properties
A delta in a streaming chat completion response.
Properties
The content added in this delta.
The refusal message added in this delta.
The role of the message author.
Variants
Tool calls made in this delta.
Items
A function tool call made by the assistant.
Properties
The index of the tool call in the sequence of tool calls.
The unique identifier of the function tool.
Properties
The name of the function.
The arguments passed to the function.
The reasoning added in this delta.
Images added in this delta.
Items
Properties
Properties
The Base64 URL of the generated image.
The reason why the assistant ceased to generate further tokens.
Variants
The index of the choice in the list of choices.
The log probabilities of the tokens generated by the model.
Properties
The log probabilities of the tokens in the content.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The log probabilities of the tokens in the refusal.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The Unix timestamp (in seconds) when the chat completion was created.
The unique identifier of the Ensemble LLM used for this chat completion.
The upstream model used for this chat completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The cost multiplier applied to upstream costs for computing ObjectiveAI costs.
Whether the completion used a BYOK (Bring Your Own Key) API Key.
The provider used for this chat completion.
When present, indicates that an error occurred during the chat completion.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
The output of the function execution.
Variants
The scalar output of the function execution.
The vector output of the function execution.
Items
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
When present, indicates that an error occurred during the function execution.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
A token which may be used to retry the function execution.
The UNIX timestamp (in seconds) when the function execution chunk was created.
The unique identifier of the function being executed.
The unique identifier of the profile being used.
The object type.
Variants
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The index of the task in the sequence of tasks.
The index of the task amongst all mapped and non-skipped compiled tasks. Used internally.
The path of this task which may be used to navigate which nested task this is amongst the root functions tasks and sub-tasks.
Items
The Swiss system round number (1-indexed).
The index of this task within its Swiss system pool.
A chunk of a vector completion task.
Properties
The unique identifier of the vector completion.
The list of chat completion chunks created for this vector completion.
Items
A chat completion chunk generated in the pursuit of a vector completion.
Properties
The unique identifier of the chat completion.
The unique identifier of the upstream chat completion.
The list of choices in this chunk.
Items
A choice in a streaming chat completion response.
Properties
A delta in a streaming chat completion response.
Properties
The content added in this delta.
The refusal message added in this delta.
The role of the message author.
Variants
Tool calls made in this delta.
Items
A function tool call made by the assistant.
Properties
The index of the tool call in the sequence of tool calls.
The unique identifier of the function tool.
Properties
The name of the function.
The arguments passed to the function.
The reasoning added in this delta.
Images added in this delta.
Items
Properties
Properties
The Base64 URL of the generated image.
The reason why the assistant ceased to generate further tokens.
Variants
The index of the choice in the list of choices.
The log probabilities of the tokens generated by the model.
Properties
The log probabilities of the tokens in the content.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The log probabilities of the tokens in the refusal.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The Unix timestamp (in seconds) when the chat completion was created.
The unique identifier of the Ensemble LLM used for this chat completion.
The upstream model used for this chat completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The cost multiplier applied to upstream costs for computing ObjectiveAI costs.
Whether the completion used a BYOK (Bring Your Own Key) API Key.
The provider used for this chat completion.
The index of the completion amongst all chat completions.
An error encountered during the generation of this chat completion.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
The list of votes for responses in the request from the Ensemble LLMs within the provided Ensemble.
Items
A vote from an Ensemble LLM within a Vector Completion.
Properties
The unique identifier of the Ensemble LLM which generated this vote.
The index of the Ensemble LLM in the Ensemble.
The flat index of the Ensemble LLM in the expanded Ensemble, accounting for counts.
The vote generated by this Ensemble LLM. It is of the same length of the number of responses provided in the request. If the Ensemble LLM used logprobs, may be a probability distribution; otherwise, one of the responses will have a value of 1 and the rest 0.
Items
The weight assigned to this vote.
Whether this vote came from a previous Vector Completion which was retried. from_cache will also be true.
Whether this vote came from the global ObjectiveAI votes cache.
Whether this vote was generated via RNG.
The scores for each response in the request, aggregated from the votes of the Ensemble LLMs.
Items
The weights assigned to each response in the request, aggregated from the votes of the Ensemble LLMs.
Items
The Unix timestamp (in seconds) when the vector completion was created.
The unique identifier of the Ensemble used for this vector completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The index of the task in the sequence of tasks.
The index of the task amongst all mapped and non-skipped compiled tasks. Used internally.
The path of this task which may be used to navigate which nested task this is amongst the root functions tasks and sub-tasks.
Items
When present, indicates that an error occurred during the vector completion task.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
When true, indicates that one or more tasks encountered errors during execution.
A chunk of a reasoning summary generation.
Properties
The unique identifier of the chat completion.
The unique identifier of the upstream chat completion.
The list of choices in this chunk.
Items
A choice in a streaming chat completion response.
Properties
A delta in a streaming chat completion response.
Properties
The content added in this delta.
The refusal message added in this delta.
The role of the message author.
Variants
Tool calls made in this delta.
Items
A function tool call made by the assistant.
Properties
The index of the tool call in the sequence of tool calls.
The unique identifier of the function tool.
Properties
The name of the function.
The arguments passed to the function.
The reasoning added in this delta.
Images added in this delta.
Items
Properties
Properties
The Base64 URL of the generated image.
The reason why the assistant ceased to generate further tokens.
Variants
The index of the choice in the list of choices.
The log probabilities of the tokens generated by the model.
Properties
The log probabilities of the tokens in the content.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The log probabilities of the tokens in the refusal.
Items
The token which was selected by the sampler for this position as well as the logprobabilities of the top options.
Properties
The token string which was selected by the sampler.
The byte representation of the token which was selected by the sampler.
Items
The log probability of the token which was selected by the sampler.
The log probabilities of the top tokens for this position.
Items
The log probability of a token in the list of top tokens.
Properties
The token string.
The byte representation of the token.
Items
The log probability of the token.
The Unix timestamp (in seconds) when the chat completion was created.
The unique identifier of the Ensemble LLM used for this chat completion.
The upstream model used for this chat completion.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The cost multiplier applied to upstream costs for computing ObjectiveAI costs.
Whether the completion used a BYOK (Bring Your Own Key) API Key.
The provider used for this chat completion.
When present, indicates that an error occurred during the chat completion.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
The output of the function execution.
Variants
The scalar output of the function execution.
The vector output of the function execution.
Items
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
When present, indicates that an error occurred during the function execution.
Properties
The status code of the error.
The message or details of the error.
Variants
Null
Boolean
Number
String
Array
Items
A JSON value.
Object
Values
A JSON value.
A token which may be used to retry the function execution.
The UNIX timestamp (in seconds) when the function execution chunk was created.
The unique identifier of the function being executed.
The unique identifier of the profile being used.
The object type.
Variants
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.
The index of the function execution chunk in the list of executions.
The index of the dataset item this function execution chunk corresponds to.
The N index for this function execution chunk. There will be N function executions, and N comes from the request parameters.
The retry index for this function execution chunk. There may be multiple retries for a given dataset item and N index.
When true, indicates that one or more function executions encountered errors during profile computation.
A function profile defined inline.
Properties
The list of task profiles.
Items
The identifiers for a function profile hosted in a GitHub repository.
Properties
The owner of the GitHub repository containing the profile.
The name of the GitHub repository containing the profile.
The commit SHA of the GitHub repository containing the profile.
A function profile defined inline.
Properties
The list of task profiles.
Items
A vector completion profile for a vector completion task containing an Ensemble and array of weights.
Properties
The Ensemble to use for this completion. May be a unique ID or an inline definition.
Variants
The unique identifier for the Ensemble.
An ensemble of LLMs.
Properties
The list of LLMs that make up the ensemble.
Items
An LLM to be used within an Ensemble, including optional fallbacks and count.
Properties
The full ID of the LLM to use.
For Vector Completions only, specifies the LLM's voting output mode. For "instruction", the assistant is instructed to output a key. For "json_schema", the assistant is constrained to output a valid key using a JSON schema. For "tool_call", the assistant is instructed to output a tool call to select the key.
Variants
For Vector Completions only, whether to use synthetic reasoning prior to voting. Works for any LLM, even those that do not have native reasoning capabilities.
For Vector Completions only, whether to use logprobs to make the vote probabilistic. This means that the LLM can vote for multiple keys based on their logprobabilities. Allows LLMs to express native uncertainty when voting.
A list of messages exchanged in a chat conversation. These will be prepended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
A list of messages exchanged in a chat conversation. These will be appended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Values
An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
This setting aims to control the presence of tokens in the output. It tries to encourage the model to use tokens that are less present in the input, proportional to their presence in the input. Token presence scales with the number of occurrences. Negative values will encourage more diverse token usage.
The assistant will stop when any of the provided strings are generated.
Variants
Generation will stop when this string is generated.
Generation will stop when any of these strings are generated.
Items
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Options for selecting the upstream provider of this model.
Properties
Whether to allow fallback providers if the preferred provider is unavailable.
Whether to require that the provider supports all specified parameters.
An ordered list of provider names to use when selecting a provider for this model.
Items
A list of provider names to restrict selection to when selecting a provider for this model.
Items
A list of provider names to ignore when selecting a provider for this model.
Items
Specifies the quantizations to allow when selecting providers for this model.
Items
Options for controlling reasoning behavior of the model.
Properties
Enables or disables reasoning for supported models.
The maximum number of tokens to use for reasoning in a response.
Constrains effort on reasoning for supported reasoning models. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
Variants
Controls the verbosity of the reasoning summary for supported reasoning models.
Variants
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.
Variants
A count greater than one effectively means that there are multiple instances of this LLM in an ensemble.
A list of fallback LLMs to use if the primary LLM fails.
Items
An LLM to be used within an Ensemble or standalone with Chat Completions.
Properties
The full ID of the LLM to use.
For Vector Completions only, specifies the LLM's voting output mode. For "instruction", the assistant is instructed to output a key. For "json_schema", the assistant is constrained to output a valid key using a JSON schema. For "tool_call", the assistant is instructed to output a tool call to select the key.
Variants
For Vector Completions only, whether to use synthetic reasoning prior to voting. Works for any LLM, even those that do not have native reasoning capabilities.
For Vector Completions only, whether to use logprobs to make the vote probabilistic. This means that the LLM can vote for multiple keys based on their logprobabilities. Allows LLMs to express native uncertainty when voting.
A list of messages exchanged in a chat conversation. These will be prepended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
A list of messages exchanged in a chat conversation. These will be appended to every prompt sent to this LLM. Useful for setting context or influencing behavior.
Items
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Developer-provided instructions that the model should follow, regardless of messages sent by the user.
Properties
Simple content.
Variants
Plain text content.
An array of simple content parts.
Items
A simple content part.
Properties
The text content.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by an end user, containing prompts or additional context information.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
Messages sent by tools in response to tool calls made by the assistant.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
The ID of the tool call that this message is responding to.
Messages sent by the model in response to user messages.
Properties
Rich content.
Variants
Plain text content.
An array of rich content parts.
Items
A text rich content part.
Properties
The text content.
An image rich content part.
Properties
The URL of the image and its optional detail level.
Properties
Either a URL of the image or the base64 encoded image data.
Specifies the detail level of the image.
Variants
An audio rich content part.
Properties
The audio data and its format.
Properties
Base64 encoded audio data.
The format of the encoded audio data.
Variants
A video rich content part.
Properties
Variants
Properties
URL of the video.
A file rich content part.
Properties
The file to be used as input, either as base64 data, an uploaded file ID, or a URL.
Properties
The base64 encoded file data, used when passing the file to the model as a string.
The ID of an uploaded file to use as input.
The name of the file, used when passing the file to the model as a string.
The URL of the file, used when passing the file to the model as a URL.
An optional name for the participant. Provides the model information to differentiate between participants of the same role.
The refusal message by the assistant.
Tool calls made by the assistant.
Items
A function tool call made by the assistant.
Properties
The unique identifier for the tool call.
The name and arguments of the function called.
Properties
The name of the function called.
The arguments passed to the function.
The reasoning provided by the assistant.
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Values
An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
This setting aims to control the presence of tokens in the output. It tries to encourage the model to use tokens that are less present in the input, proportional to their presence in the input. Token presence scales with the number of occurrences. Negative values will encourage more diverse token usage.
The assistant will stop when any of the provided strings are generated.
Variants
Generation will stop when this string is generated.
Generation will stop when any of these strings are generated.
Items
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Options for selecting the upstream provider of this model.
Properties
Whether to allow fallback providers if the preferred provider is unavailable.
Whether to require that the provider supports all specified parameters.
An ordered list of provider names to use when selecting a provider for this model.
Items
A list of provider names to restrict selection to when selecting a provider for this model.
Items
A list of provider names to ignore when selecting a provider for this model.
Items
Specifies the quantizations to allow when selecting providers for this model.
Items
Options for controlling reasoning behavior of the model.
Properties
Enables or disables reasoning for supported models.
The maximum number of tokens to use for reasoning in a response.
Constrains effort on reasoning for supported reasoning models. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
Variants
Controls the verbosity of the reasoning summary for supported reasoning models.
Variants
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.
Variants
The profile to use for the completion. Must be of the same length as the Ensemble's llms field, ignoring count.
Items
Statistics about the fitting process used to compute the weights for the profile.
Properties
The final sum loss achieved during weights fitting.
The total number of function executions used during weights fitting.
The number of fitting starts attempted. Each start begins with a randomized weight vector.
The number of fitting rounds performed across all starts.
The number of errors which occurred while computing outputs during fitting.
A token which may be used to retry the function profile computation.
The UNIX timestamp (in seconds) when the function profile computation was created.
The unique identifier of the function for which the profile is being computed.
Token and cost usage statistics for the completion.
Properties
The number of tokens generated in the completion.
The number of tokens in the prompt.
The total number of tokens used in the prompt or generated in the completion.
Detailed breakdown of generated completion tokens.
Properties
The number of accepted prediction tokens in the completion.
The number of generated audio tokens in the completion.
The number of generated reasoning tokens in the completion.
The number of rejected prediction tokens in the completion.
Detailed breakdown of prompt tokens.
Properties
The number of audio tokens in the prompt.
The number of cached tokens in the prompt.
The number of prompt tokens written to cache.
The number of video tokens in the prompt.
The cost in credits incurred for this completion.
Detailed breakdown of upstream costs incurred.
Properties
The cost incurred upstream.
The cost incurred by upstream's upstream.
The total cost in credits incurred including upstream costs.