Docs

Request Headers

authorizationstring*required

Authorization token (required).

Request Body

ChatCompletionCreateParamsStreamingobject

Parameters for creating a streaming chat completion.

Properties
messagesarray*required

A list of messages exchanged in a chat conversation.

Items
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

providerobjectoptional

Options for selecting the upstream provider of this completion.

Properties
data_collectionenumoptional

Specifies whether to allow providers which collect data.

Variants
"allow"
"deny"
zdrbooleanoptional

Whether to enforce Zero Data Retention (ZDR) policies when selecting providers.

sortenumoptional

Specifies the sorting strategy for provider selection.

Variants
"price"
"throughput"
"latency"
max_priceobjectoptional
Properties
promptnumberoptional

Maximum price for prompt tokens.

completionnumberoptional

Maximum price for completion tokens.

imagenumberoptional

Maximum price for image generation.

audionumberoptional

Maximum price for audio generation.

requestnumberoptional

Maximum price per request.

preferred_min_throughputnumberoptional

Preferred minimum throughput for the provider.

preferred_max_latencynumberoptional

Preferred maximum latency for the provider.

min_throughputnumberoptional

Minimum throughput for the provider.

max_latencynumberoptional

Maximum latency for the provider.

modelunion*required

The Ensemble LLM to use for this completion. May be a unique ID or an inline definition.

Variants
string
object

An LLM to be used within an Ensemble or standalone with Chat Completions.

Properties
modelstring*required

The full ID of the LLM to use.

output_modeenum*required

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
"instruction"
"json_schema"
"tool_call"
synthetic_reasoningbooleanoptional

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.

top_logprobsnumbermin: 0max: 20optional

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.

prefix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

suffix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

frequency_penaltynumbermin: -2max: 2optional

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.

logit_biasrecordoptional

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
numbermin: -100max: 100
max_completion_tokensnumbermin: 0max: 2147483647optional

An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.

presence_penaltynumbermin: -2max: 2optional

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.

stopunionoptional

The assistant will stop when any of the provided strings are generated.

Variants
StopStringstring

Generation will stop when this string is generated.

StopStringsarray

Generation will stop when any of these strings are generated.

Items
StopStringstring
temperaturenumbermin: 0max: 2optional

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.

top_pnumbermin: 0max: 1optional

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.

max_tokensnumbermin: 0max: 2147483647optional

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.

min_pnumbermin: 0max: 1optional

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.

providerobjectoptional

Options for selecting the upstream provider of this model.

Properties
allow_fallbacksbooleanoptional

Whether to allow fallback providers if the preferred provider is unavailable.

require_parametersbooleanoptional

Whether to require that the provider supports all specified parameters.

orderarrayoptional

An ordered list of provider names to use when selecting a provider for this model.

Items
ProviderNamestring
onlyarrayoptional

A list of provider names to restrict selection to when selecting a provider for this model.

Items
ProviderNamestring
ignorearrayoptional

A list of provider names to ignore when selecting a provider for this model.

Items
ProviderNamestring
quantizationsarrayoptional

Specifies the quantizations to allow when selecting providers for this model.

Items
"int4"
"int8"
"fp4"
"fp6"
"fp8"
"fp16"
"bf16"
"fp32"
"unknown"
reasoningobjectoptional

Options for controlling reasoning behavior of the model.

Properties
enabledbooleanoptional

Enables or disables reasoning for supported models.

max_tokensnumbermin: 0max: 2147483647optional

The maximum number of tokens to use for reasoning in a response.

effortenumoptional

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
"none"
"minimal"
"low"
"medium"
"high"
"xhigh"
summary_verbosityenumoptional

Controls the verbosity of the reasoning summary for supported reasoning models.

Variants
"auto"
"concise"
"detailed"
repetition_penaltynumbermin: 0max: 2optional

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.

top_anumbermin: 0max: 1optional

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.

top_knumbermin: 0max: 2147483647optional

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.

verbosityenumoptional

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
"low"
"medium"
"high"
modelsarrayoptional

Fallback Ensemble LLMs to use if the primary Ensemble LLM fails.

Items
string
object

An LLM to be used within an Ensemble or standalone with Chat Completions.

Properties
modelstring*required

The full ID of the LLM to use.

output_modeenum*required

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
"instruction"
"json_schema"
"tool_call"
synthetic_reasoningbooleanoptional

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.

top_logprobsnumbermin: 0max: 20optional

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.

prefix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

suffix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

frequency_penaltynumbermin: -2max: 2optional

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.

logit_biasrecordoptional

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
numbermin: -100max: 100
max_completion_tokensnumbermin: 0max: 2147483647optional

An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.

presence_penaltynumbermin: -2max: 2optional

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.

stopunionoptional

The assistant will stop when any of the provided strings are generated.

Variants
StopStringstring

Generation will stop when this string is generated.

StopStringsarray

Generation will stop when any of these strings are generated.

Items
StopStringstring
temperaturenumbermin: 0max: 2optional

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.

top_pnumbermin: 0max: 1optional

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.

max_tokensnumbermin: 0max: 2147483647optional

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.

min_pnumbermin: 0max: 1optional

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.

providerobjectoptional

Options for selecting the upstream provider of this model.

Properties
allow_fallbacksbooleanoptional

Whether to allow fallback providers if the preferred provider is unavailable.

require_parametersbooleanoptional

Whether to require that the provider supports all specified parameters.

orderarrayoptional

An ordered list of provider names to use when selecting a provider for this model.

Items
ProviderNamestring
onlyarrayoptional

A list of provider names to restrict selection to when selecting a provider for this model.

Items
ProviderNamestring
ignorearrayoptional

A list of provider names to ignore when selecting a provider for this model.

Items
ProviderNamestring
quantizationsarrayoptional

Specifies the quantizations to allow when selecting providers for this model.

Items
"int4"
"int8"
"fp4"
"fp6"
"fp8"
"fp16"
"bf16"
"fp32"
"unknown"
reasoningobjectoptional

Options for controlling reasoning behavior of the model.

Properties
enabledbooleanoptional

Enables or disables reasoning for supported models.

max_tokensnumbermin: 0max: 2147483647optional

The maximum number of tokens to use for reasoning in a response.

effortenumoptional

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
"none"
"minimal"
"low"
"medium"
"high"
"xhigh"
summary_verbosityenumoptional

Controls the verbosity of the reasoning summary for supported reasoning models.

Variants
"auto"
"concise"
"detailed"
repetition_penaltynumbermin: 0max: 2optional

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.

top_anumbermin: 0max: 1optional

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.

top_knumbermin: 0max: 2147483647optional

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.

verbosityenumoptional

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
"low"
"medium"
"high"
top_logprobsnumbermin: 0max: 20optional

An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability.

response_formatunionoptional

The desired format of the model's response.

Variants
ResponseFormatTextobject

The response will be arbitrary text.

Properties
type"text"*required
ResponseFormatJsonObjectobject

The response will be a JSON object.

Properties
type"json_object"*required
ResponseFormatJsonSchemaobject

The response will conform to the provided JSON schema.

Properties
type"json_schema"*required
json_schemaobject*required

A JSON schema definition for constraining model output.

Properties
namestring*required

The name of the JSON schema.

descriptionstringoptional

The description of the JSON schema.

schemaanyoptional

The JSON schema definition.

strictbooleanoptional

Whether to enforce strict adherence to the JSON schema.

ResponseFormatGrammarobject

The response will conform to the provided grammar definition.

Properties
type"grammar"*required
grammarstring*required

The grammar definition to constrain the response.

ResponseFormatPythonobject

The response will be Python code.

Properties
type"python"*required
seedbigintoptional

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.

tool_choiceunionoptional

Specifies tool call behavior for the assistant.

Variants
"none"
"auto"
"required"
ToolChoiceFunctionobject

Specify a function for the assistant to call.

Properties
type"function"*required
functionobject*required
Properties
namestring*required

The name of the function the assistant will call.

toolsarray*required

A list of tools that the assistant can call.

Items
FunctionToolobject

A function tool that the assistant can call.

Properties
type"function"*required
functionobject*required

The definition of a function tool.

Properties
namestring*required

The name of the function.

descriptionstringoptional

The description of the function.

parametersrecordoptional

The JSON schema defining the parameters of the function.

Values
null
boolean
number
string
array
Items
JsonValue

A JSON value.

record
Values
JsonValue

A JSON value.

strictbooleanoptional

Whether to enforce strict adherence to the parameter schema.

parallel_tool_callsbooleanoptional

Whether to allow the model to make multiple tool calls in parallel.

predictionobjectoptional

Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time. This is most common when you are regenerating a file with only minor changes to most of the content.

Properties
type"content"*required
contentunion*required
Variants
PredictionContentTextstring
PredictionContentPartsarray
Items
PredictionContentPartobject

A part of the predicted content.

Properties
type"text"*required
textstring*required
backoff_max_elapsed_timenumbermin: 0max: 4294967295optional

The maximum total time in milliseconds to spend on retries when a transient error occurs.

first_chunk_timeoutnumbermin: 0max: 4294967295optional

The maximum time in milliseconds to wait for the first chunk of a streaming response.

other_chunk_timeoutnumbermin: 0max: 4294967295optional

The maximum time in milliseconds to wait between subsequent chunks of a streaming response.

stream"true"*required

Whether to stream the response as a series of chunks.

ChatCompletionCreateParamsNonStreamingobject

Parameters for creating a unary chat completion.

Properties
messagesarray*required

A list of messages exchanged in a chat conversation.

Items
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

providerobjectoptional

Options for selecting the upstream provider of this completion.

Properties
data_collectionenumoptional

Specifies whether to allow providers which collect data.

Variants
"allow"
"deny"
zdrbooleanoptional

Whether to enforce Zero Data Retention (ZDR) policies when selecting providers.

sortenumoptional

Specifies the sorting strategy for provider selection.

Variants
"price"
"throughput"
"latency"
max_priceobjectoptional
Properties
promptnumberoptional

Maximum price for prompt tokens.

completionnumberoptional

Maximum price for completion tokens.

imagenumberoptional

Maximum price for image generation.

audionumberoptional

Maximum price for audio generation.

requestnumberoptional

Maximum price per request.

preferred_min_throughputnumberoptional

Preferred minimum throughput for the provider.

preferred_max_latencynumberoptional

Preferred maximum latency for the provider.

min_throughputnumberoptional

Minimum throughput for the provider.

max_latencynumberoptional

Maximum latency for the provider.

modelunion*required

The Ensemble LLM to use for this completion. May be a unique ID or an inline definition.

Variants
string
object

An LLM to be used within an Ensemble or standalone with Chat Completions.

Properties
modelstring*required

The full ID of the LLM to use.

output_modeenum*required

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
"instruction"
"json_schema"
"tool_call"
synthetic_reasoningbooleanoptional

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.

top_logprobsnumbermin: 0max: 20optional

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.

prefix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

suffix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

frequency_penaltynumbermin: -2max: 2optional

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.

logit_biasrecordoptional

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
numbermin: -100max: 100
max_completion_tokensnumbermin: 0max: 2147483647optional

An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.

presence_penaltynumbermin: -2max: 2optional

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.

stopunionoptional

The assistant will stop when any of the provided strings are generated.

Variants
StopStringstring

Generation will stop when this string is generated.

StopStringsarray

Generation will stop when any of these strings are generated.

Items
StopStringstring
temperaturenumbermin: 0max: 2optional

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.

top_pnumbermin: 0max: 1optional

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.

max_tokensnumbermin: 0max: 2147483647optional

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.

min_pnumbermin: 0max: 1optional

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.

providerobjectoptional

Options for selecting the upstream provider of this model.

Properties
allow_fallbacksbooleanoptional

Whether to allow fallback providers if the preferred provider is unavailable.

require_parametersbooleanoptional

Whether to require that the provider supports all specified parameters.

orderarrayoptional

An ordered list of provider names to use when selecting a provider for this model.

Items
ProviderNamestring
onlyarrayoptional

A list of provider names to restrict selection to when selecting a provider for this model.

Items
ProviderNamestring
ignorearrayoptional

A list of provider names to ignore when selecting a provider for this model.

Items
ProviderNamestring
quantizationsarrayoptional

Specifies the quantizations to allow when selecting providers for this model.

Items
"int4"
"int8"
"fp4"
"fp6"
"fp8"
"fp16"
"bf16"
"fp32"
"unknown"
reasoningobjectoptional

Options for controlling reasoning behavior of the model.

Properties
enabledbooleanoptional

Enables or disables reasoning for supported models.

max_tokensnumbermin: 0max: 2147483647optional

The maximum number of tokens to use for reasoning in a response.

effortenumoptional

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
"none"
"minimal"
"low"
"medium"
"high"
"xhigh"
summary_verbosityenumoptional

Controls the verbosity of the reasoning summary for supported reasoning models.

Variants
"auto"
"concise"
"detailed"
repetition_penaltynumbermin: 0max: 2optional

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.

top_anumbermin: 0max: 1optional

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.

top_knumbermin: 0max: 2147483647optional

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.

verbosityenumoptional

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
"low"
"medium"
"high"
modelsarrayoptional

Fallback Ensemble LLMs to use if the primary Ensemble LLM fails.

Items
string
object

An LLM to be used within an Ensemble or standalone with Chat Completions.

Properties
modelstring*required

The full ID of the LLM to use.

output_modeenum*required

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
"instruction"
"json_schema"
"tool_call"
synthetic_reasoningbooleanoptional

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.

top_logprobsnumbermin: 0max: 20optional

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.

prefix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

suffix_messagesarrayoptional

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
DeveloperMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"developer"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

SystemMessageobject

Developer-provided instructions that the model should follow, regardless of messages sent by the user.

Properties
role"system"*required
contentunion*required

Simple content.

Variants
SimpleContentTextstring

Plain text content.

SimpleContentPartsarray

An array of simple content parts.

Items
SimpleContentPartobject

A simple content part.

Properties
type"text"*required
textstring*required

The text content.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

UserMessageobject

Messages sent by an end user, containing prompts or additional context information.

Properties
role"user"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

ToolMessageobject

Messages sent by tools in response to tool calls made by the assistant.

Properties
role"tool"*required
contentunion*required

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

tool_call_idstring*required

The ID of the tool call that this message is responding to.

AssistantMessageobject

Messages sent by the model in response to user messages.

Properties
role"assistant"*required
contentunionoptional

Rich content.

Variants
RichContentTextstring

Plain text content.

RichContentPartsarray

An array of rich content parts.

Items
TextRichContentPartobject

A text rich content part.

Properties
type"text"*required
textstring*required

The text content.

ImageRichContentPartobject

An image rich content part.

Properties
type"image_url"*required
image_urlobject*required

The URL of the image and its optional detail level.

Properties
urlstring*required

Either a URL of the image or the base64 encoded image data.

detailenumoptional

Specifies the detail level of the image.

Variants
"auto"
"low"
"high"
AudioRichContentPartobject

An audio rich content part.

Properties
type"input_audio"*required
input_audioobject*required

The audio data and its format.

Properties
datastring*required

Base64 encoded audio data.

formatenum*required

The format of the encoded audio data.

Variants
"wav"
"mp3"
VideoRichContentPartobject

A video rich content part.

Properties
typeenum*required
Variants
"video_url"
"input_video"
video_urlobject*required
Properties
urlstring*required

URL of the video.

FileRichContentPartobject

A file rich content part.

Properties
type"file"*required
fileobject*required

The file to be used as input, either as base64 data, an uploaded file ID, or a URL.

Properties
file_datastringoptional

The base64 encoded file data, used when passing the file to the model as a string.

file_idstringoptional

The ID of an uploaded file to use as input.

filenamestringoptional

The name of the file, used when passing the file to the model as a string.

file_urlstringoptional

The URL of the file, used when passing the file to the model as a URL.

namestringoptional

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

refusalstringoptional

The refusal message by the assistant.

tool_callsarrayoptional

Tool calls made by the assistant.

Items
AssistantMessageToolCallFunctionobject

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier for the tool call.

functionobject*required

The name and arguments of the function called.

Properties
namestring*required

The name of the function called.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant.

frequency_penaltynumbermin: -2max: 2optional

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.

logit_biasrecordoptional

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
numbermin: -100max: 100
max_completion_tokensnumbermin: 0max: 2147483647optional

An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.

presence_penaltynumbermin: -2max: 2optional

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.

stopunionoptional

The assistant will stop when any of the provided strings are generated.

Variants
StopStringstring

Generation will stop when this string is generated.

StopStringsarray

Generation will stop when any of these strings are generated.

Items
StopStringstring
temperaturenumbermin: 0max: 2optional

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.

top_pnumbermin: 0max: 1optional

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.

max_tokensnumbermin: 0max: 2147483647optional

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.

min_pnumbermin: 0max: 1optional

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.

providerobjectoptional

Options for selecting the upstream provider of this model.

Properties
allow_fallbacksbooleanoptional

Whether to allow fallback providers if the preferred provider is unavailable.

require_parametersbooleanoptional

Whether to require that the provider supports all specified parameters.

orderarrayoptional

An ordered list of provider names to use when selecting a provider for this model.

Items
ProviderNamestring
onlyarrayoptional

A list of provider names to restrict selection to when selecting a provider for this model.

Items
ProviderNamestring
ignorearrayoptional

A list of provider names to ignore when selecting a provider for this model.

Items
ProviderNamestring
quantizationsarrayoptional

Specifies the quantizations to allow when selecting providers for this model.

Items
"int4"
"int8"
"fp4"
"fp6"
"fp8"
"fp16"
"bf16"
"fp32"
"unknown"
reasoningobjectoptional

Options for controlling reasoning behavior of the model.

Properties
enabledbooleanoptional

Enables or disables reasoning for supported models.

max_tokensnumbermin: 0max: 2147483647optional

The maximum number of tokens to use for reasoning in a response.

effortenumoptional

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
"none"
"minimal"
"low"
"medium"
"high"
"xhigh"
summary_verbosityenumoptional

Controls the verbosity of the reasoning summary for supported reasoning models.

Variants
"auto"
"concise"
"detailed"
repetition_penaltynumbermin: 0max: 2optional

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.

top_anumbermin: 0max: 1optional

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.

top_knumbermin: 0max: 2147483647optional

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.

verbosityenumoptional

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
"low"
"medium"
"high"
top_logprobsnumbermin: 0max: 20optional

An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability.

response_formatunionoptional

The desired format of the model's response.

Variants
ResponseFormatTextobject

The response will be arbitrary text.

Properties
type"text"*required
ResponseFormatJsonObjectobject

The response will be a JSON object.

Properties
type"json_object"*required
ResponseFormatJsonSchemaobject

The response will conform to the provided JSON schema.

Properties
type"json_schema"*required
json_schemaobject*required

A JSON schema definition for constraining model output.

Properties
namestring*required

The name of the JSON schema.

descriptionstringoptional

The description of the JSON schema.

schemaanyoptional

The JSON schema definition.

strictbooleanoptional

Whether to enforce strict adherence to the JSON schema.

ResponseFormatGrammarobject

The response will conform to the provided grammar definition.

Properties
type"grammar"*required
grammarstring*required

The grammar definition to constrain the response.

ResponseFormatPythonobject

The response will be Python code.

Properties
type"python"*required
seedbigintoptional

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.

tool_choiceunionoptional

Specifies tool call behavior for the assistant.

Variants
"none"
"auto"
"required"
ToolChoiceFunctionobject

Specify a function for the assistant to call.

Properties
type"function"*required
functionobject*required
Properties
namestring*required

The name of the function the assistant will call.

toolsarray*required

A list of tools that the assistant can call.

Items
FunctionToolobject

A function tool that the assistant can call.

Properties
type"function"*required
functionobject*required

The definition of a function tool.

Properties
namestring*required

The name of the function.

descriptionstringoptional

The description of the function.

parametersrecordoptional

The JSON schema defining the parameters of the function.

Values
null
boolean
number
string
array
Items
JsonValue

A JSON value.

record
Values
JsonValue

A JSON value.

strictbooleanoptional

Whether to enforce strict adherence to the parameter schema.

parallel_tool_callsbooleanoptional

Whether to allow the model to make multiple tool calls in parallel.

predictionobjectoptional

Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time. This is most common when you are regenerating a file with only minor changes to most of the content.

Properties
type"content"*required
contentunion*required
Variants
PredictionContentTextstring
PredictionContentPartsarray
Items
PredictionContentPartobject

A part of the predicted content.

Properties
type"text"*required
textstring*required
backoff_max_elapsed_timenumbermin: 0max: 4294967295optional

The maximum total time in milliseconds to spend on retries when a transient error occurs.

first_chunk_timeoutnumbermin: 0max: 4294967295optional

The maximum time in milliseconds to wait for the first chunk of a streaming response.

other_chunk_timeoutnumbermin: 0max: 4294967295optional

The maximum time in milliseconds to wait between subsequent chunks of a streaming response.

stream"false"optional

Whether to stream the response as a series of chunks.

Response Body

idstring*required

The unique identifier of the chat completion.

upstream_idstring*required

The unique identifier of the upstream chat completion.

choicesarray*required

The list of choices in this chat completion.

Items
object

A choice in a unary chat completion response.

Properties
messageobject*required

A message generated by the assistant.

Properties
contentstringoptional

The content of the message.

refusalstringoptional

The refusal message, if any.

roleenum*required

The role of the message author.

Variants
"assistant"
tool_callsarrayoptional

The tool calls made by the assistant, if any.

Items
object

A function tool call made by the assistant.

Properties
type"function"*required
idstring*required

The unique identifier of the function tool.

functionobject*required
Properties
namestring*required

The name of the function.

argumentsstring*required

The arguments passed to the function.

reasoningstringoptional

The reasoning provided by the assistant, if any.

imagesarrayoptional

The images generated by the assistant, if any.

Items
object
Properties
type"image_url"*required
image_urlobject*required
Properties
urlstring*required

The Base64 URL of the generated image.

finish_reasonenum*required

The reason why the assistant ceased to generate further tokens.

Variants
"stop"
"length"
"tool_calls"
"content_filter"
"error"
indexnumbermin: 0max: 4294967295*required

The index of the choice in the list of choices.

logprobsobjectoptional

The log probabilities of the tokens generated by the model.

Properties
contentarrayoptional

The log probabilities of the tokens in the content.

Items
object

The token which was selected by the sampler for this position as well as the logprobabilities of the top options.

Properties
tokenstring*required

The token string which was selected by the sampler.

bytesarrayoptional

The byte representation of the token which was selected by the sampler.

Items
numbermin: 0max: 4294967295
logprobnumber*required

The log probability of the token which was selected by the sampler.

top_logprobsarray*required

The log probabilities of the top tokens for this position.

Items
object

The log probability of a token in the list of top tokens.

Properties
tokenstring*required

The token string.

bytesarrayoptional

The byte representation of the token.

Items
numbermin: 0max: 4294967295
logprobnumberoptional

The log probability of the token.

refusalarrayoptional

The log probabilities of the tokens in the refusal.

Items
object

The token which was selected by the sampler for this position as well as the logprobabilities of the top options.

Properties
tokenstring*required

The token string which was selected by the sampler.

bytesarrayoptional

The byte representation of the token which was selected by the sampler.

Items
numbermin: 0max: 4294967295
logprobnumber*required

The log probability of the token which was selected by the sampler.

top_logprobsarray*required

The log probabilities of the top tokens for this position.

Items
object

The log probability of a token in the list of top tokens.

Properties
tokenstring*required

The token string.

bytesarrayoptional

The byte representation of the token.

Items
numbermin: 0max: 4294967295
logprobnumberoptional

The log probability of the token.

creatednumbermin: 0max: 4294967295*required

The Unix timestamp (in seconds) when the chat completion was created.

modelstring*required

The unique identifier of the Ensemble LLM used for this chat completion.

upstream_modelstring*required

The upstream model used for this chat completion.

object"chat.completion"*required
service_tierstringoptional
system_fingerprintstringoptional
usageobject*required

Token and cost usage statistics for the completion.

Properties
completion_tokensnumbermin: 0max: 4294967295*required

The number of tokens generated in the completion.

prompt_tokensnumbermin: 0max: 4294967295*required

The number of tokens in the prompt.

total_tokensnumbermin: 0max: 4294967295*required

The total number of tokens used in the prompt or generated in the completion.

completion_tokens_detailsobjectoptional

Detailed breakdown of generated completion tokens.

Properties
accepted_prediction_tokensnumbermin: 0max: 4294967295optional

The number of accepted prediction tokens in the completion.

audio_tokensnumbermin: 0max: 4294967295optional

The number of generated audio tokens in the completion.

reasoning_tokensnumbermin: 0max: 4294967295optional

The number of generated reasoning tokens in the completion.

rejected_prediction_tokensnumbermin: 0max: 4294967295optional

The number of rejected prediction tokens in the completion.

prompt_tokens_detailsobjectoptional

Detailed breakdown of prompt tokens.

Properties
audio_tokensnumbermin: 0max: 4294967295optional

The number of audio tokens in the prompt.

cached_tokensnumbermin: 0max: 4294967295optional

The number of cached tokens in the prompt.

cache_write_tokensnumbermin: 0max: 4294967295optional

The number of prompt tokens written to cache.

video_tokensnumbermin: 0max: 4294967295optional

The number of video tokens in the prompt.

costnumber*required

The cost in credits incurred for this completion.

cost_detailsobjectoptional

Detailed breakdown of upstream costs incurred.

Properties
upstream_inference_costnumberoptional

The cost incurred upstream.

upstream_upstream_inference_costnumberoptional

The cost incurred by upstream's upstream.

total_costnumber*required

The total cost in credits incurred including upstream costs.

cost_multipliernumber*required

The cost multiplier applied to upstream costs for computing ObjectiveAI costs.

is_byokboolean*required

Whether the completion used a BYOK (Bring Your Own Key) API Key.

providerstringoptional

The provider used for this chat completion.

Response Body (Streaming)

idstring*required

The unique identifier of the chat completion.

upstream_idstring*required

The unique identifier of the upstream chat completion.

choicesarray*required

The list of choices in this chunk.

Items
object

A choice in a streaming chat completion response.

Properties
deltaobject*required

A delta in a streaming chat completion response.

Properties
contentstringoptional

The content added in this delta.

refusalstringoptional

The refusal message added in this delta.

roleenumoptional

The role of the message author.

Variants
"assistant"
tool_callsarrayoptional

Tool calls made in this delta.

Items
object

A function tool call made by the assistant.

Properties
indexnumbermin: 0max: 4294967295*required

The index of the tool call in the sequence of tool calls.

type"function"optional
idstringoptional

The unique identifier of the function tool.

functionobjectoptional
Properties
namestringoptional

The name of the function.

argumentsstringoptional

The arguments passed to the function.

reasoningstringoptional

The reasoning added in this delta.

imagesarrayoptional

Images added in this delta.

Items
object
Properties
type"image_url"*required
image_urlobject*required
Properties
urlstring*required

The Base64 URL of the generated image.

finish_reasonenumoptional

The reason why the assistant ceased to generate further tokens.

Variants
"stop"
"length"
"tool_calls"
"content_filter"
"error"
indexnumbermin: 0max: 4294967295*required

The index of the choice in the list of choices.

logprobsobjectoptional

The log probabilities of the tokens generated by the model.

Properties
contentarrayoptional

The log probabilities of the tokens in the content.

Items
object

The token which was selected by the sampler for this position as well as the logprobabilities of the top options.

Properties
tokenstring*required

The token string which was selected by the sampler.

bytesarrayoptional

The byte representation of the token which was selected by the sampler.

Items
numbermin: 0max: 4294967295
logprobnumber*required

The log probability of the token which was selected by the sampler.

top_logprobsarray*required

The log probabilities of the top tokens for this position.

Items
object

The log probability of a token in the list of top tokens.

Properties
tokenstring*required

The token string.

bytesarrayoptional

The byte representation of the token.

Items
numbermin: 0max: 4294967295
logprobnumberoptional

The log probability of the token.

refusalarrayoptional

The log probabilities of the tokens in the refusal.

Items
object

The token which was selected by the sampler for this position as well as the logprobabilities of the top options.

Properties
tokenstring*required

The token string which was selected by the sampler.

bytesarrayoptional

The byte representation of the token which was selected by the sampler.

Items
numbermin: 0max: 4294967295
logprobnumber*required

The log probability of the token which was selected by the sampler.

top_logprobsarray*required

The log probabilities of the top tokens for this position.

Items
object

The log probability of a token in the list of top tokens.

Properties
tokenstring*required

The token string.

bytesarrayoptional

The byte representation of the token.

Items
numbermin: 0max: 4294967295
logprobnumberoptional

The log probability of the token.

creatednumbermin: 0max: 4294967295*required

The Unix timestamp (in seconds) when the chat completion was created.

modelstring*required

The unique identifier of the Ensemble LLM used for this chat completion.

upstream_modelstring*required

The upstream model used for this chat completion.

object"chat.completion.chunk"*required
service_tierstringoptional
system_fingerprintstringoptional
usageobjectoptional

Token and cost usage statistics for the completion.

Properties
completion_tokensnumbermin: 0max: 4294967295*required

The number of tokens generated in the completion.

prompt_tokensnumbermin: 0max: 4294967295*required

The number of tokens in the prompt.

total_tokensnumbermin: 0max: 4294967295*required

The total number of tokens used in the prompt or generated in the completion.

completion_tokens_detailsobjectoptional

Detailed breakdown of generated completion tokens.

Properties
accepted_prediction_tokensnumbermin: 0max: 4294967295optional

The number of accepted prediction tokens in the completion.

audio_tokensnumbermin: 0max: 4294967295optional

The number of generated audio tokens in the completion.

reasoning_tokensnumbermin: 0max: 4294967295optional

The number of generated reasoning tokens in the completion.

rejected_prediction_tokensnumbermin: 0max: 4294967295optional

The number of rejected prediction tokens in the completion.

prompt_tokens_detailsobjectoptional

Detailed breakdown of prompt tokens.

Properties
audio_tokensnumbermin: 0max: 4294967295optional

The number of audio tokens in the prompt.

cached_tokensnumbermin: 0max: 4294967295optional

The number of cached tokens in the prompt.

cache_write_tokensnumbermin: 0max: 4294967295optional

The number of prompt tokens written to cache.

video_tokensnumbermin: 0max: 4294967295optional

The number of video tokens in the prompt.

costnumber*required

The cost in credits incurred for this completion.

cost_detailsobjectoptional

Detailed breakdown of upstream costs incurred.

Properties
upstream_inference_costnumberoptional

The cost incurred upstream.

upstream_upstream_inference_costnumberoptional

The cost incurred by upstream's upstream.

total_costnumber*required

The total cost in credits incurred including upstream costs.

cost_multipliernumber*required

The cost multiplier applied to upstream costs for computing ObjectiveAI costs.

is_byokboolean*required

Whether the completion used a BYOK (Bring Your Own Key) API Key.

providerstringoptional

The provider used for this chat completion.

Objective Artificial Intelligence, Inc.