Chat Completions - Cerebras Inference

from cerebras.cloud.sdk import Cerebras
import os 

client = Cerebras(api_key=os.environ.get("CEREBRAS_API_KEY"),)

chat_completion = client.chat.completions.create(
    model="gpt-oss-120b",
    messages=[
        {"role": "user", "content": "Hello!",}
    ],
)
print(chat_completion)

{
  "id": "chatcmpl-30b3c3d8-ca41-48e7-9ef0-27e322604a13",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! 👋 How can I help you today?",
        "reasoning": "The user just says \"Hello!\" with no further context.\n\nWe need to respond politely, maybe ask how can assist.\n\nWe can also be friendly.\n\nNo constraints. We'll respond as chat.",
        "tool_calls": null
      },
      "logprobs": null,
      "reasoning_logprobs": null
    }
  ],
  "created": 1769729480,
  "model": "gpt-oss-120b",
  "object": "chat.completion",
  "system_fingerprint": "fp_e7ab83753cbd28777b40",
  "time_info": {
    "completion_time": 0.04040406,
    "prompt_time": 0.00383762,
    "queue_time": 0.004628115,
    "total_time": 0.05013155937194824,
    "created": 1769729480.0787008
  },
  "usage": {
    "completion_tokens": 59,
    "completion_tokens_details": {
      "accepted_prediction_tokens": 0,
      "rejected_prediction_tokens": 0,
      "reasoning_tokens": 0
    },
    "prompt_tokens": 69,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "total_tokens": 128
  },
  "service_tier": null
}

POST

chat

completions

from cerebras.cloud.sdk import Cerebras
import os 

client = Cerebras(api_key=os.environ.get("CEREBRAS_API_KEY"),)

chat_completion = client.chat.completions.create(
    model="gpt-oss-120b",
    messages=[
        {"role": "user", "content": "Hello!",}
    ],
)
print(chat_completion)

{
  "id": "chatcmpl-30b3c3d8-ca41-48e7-9ef0-27e322604a13",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! 👋 How can I help you today?",
        "reasoning": "The user just says \"Hello!\" with no further context.\n\nWe need to respond politely, maybe ask how can assist.\n\nWe can also be friendly.\n\nNo constraints. We'll respond as chat.",
        "tool_calls": null
      },
      "logprobs": null,
      "reasoning_logprobs": null
    }
  ],
  "created": 1769729480,
  "model": "gpt-oss-120b",
  "object": "chat.completion",
  "system_fingerprint": "fp_e7ab83753cbd28777b40",
  "time_info": {
    "completion_time": 0.04040406,
    "prompt_time": 0.00383762,
    "queue_time": 0.004628115,
    "total_time": 0.05013155937194824,
    "created": 1769729480.0787008
  },
  "usage": {
    "completion_tokens": 59,
    "completion_tokens_details": {
      "accepted_prediction_tokens": 0,
      "rejected_prediction_tokens": 0,
      "reasoning_tokens": 0
    },
    "prompt_tokens": 69,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "total_tokens": 128
  },
  "service_tier": null
}

Generate conversational responses using a structured message format with roles (system, user, assistant, developer, tool). Best for chatbots, assistants, and multi-turn conversations. Parameter support can differ depending on the model used to generate the response, particularly for newer reasoning models. For details about parameters in reasoning models, refer to the Reasoning Guide.

Request

Headers

Content-Type

string

The media type of the request body.Supported values: application/json, application/vnd.msgpackDefault: application/jsonSee Payload Optimization for details.

Content-Encoding

string

The compression encoding applied to the request body.Supported values: gzipWhen set, the request body must be gzip-compressed. Can be combined with any supported Content-Type.See Payload Optimization for details.

queue_threshold

string

Controls the queue time threshold for requests using the flex or auto service tiers. Requests are preemptively rejected if the rolling average queue time exceeds this threshold.

This feature is in Private Preview. For access or more information, contact us or reach out to your account representative.

Valid range: 50 - 20000 (milliseconds)Default: System default if not specifiedSee Service Tiers for more information.

Body

messages

object[]

required

A list of messages comprising the conversation so far.

Show possible types

Assistant message `object`

Messages sent by the model in response to user messages.

Show properties

content

string | array | null

The contents of the assistant message.

Show possible types

Text content `string`

The contents of the assistant message.

Array of content parts `array`

An array of content parts with a defined type. For assistant messages, only type text is supported.

Show possible types

text

string

required

The text content.

type

string

required

The type of content part. Always text.

role

string

required

The role of the messages author, in this case assistant.

name

string

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

reasoning

string | null

The reasoning content from the model’s response. Include this when passing back assistant messages that contained reasoning.

tool_calls

array

The tool calls generated by the model, such as function calls.

Show properties

string

required

The ID of the tool call.

type

string

required

The type of the tool. Currently, only function is supported.

function

object

required

The function that the model called.

Show properties

name

string

required

The name of the function to call.

arguments

string

required

The arguments to call the function with, as generated by the model in JSON format.

Developer message `object`

Developer-provided instructions that the model should follow, regardless of messages sent by the user. For gpt-oss-120b, developer messages replace the previous system messages, but system is still accepted.

The developer role is currently only available for the gpt-oss-120b model.

Show properties

content

string | array

required

The contents of the developer message.

Show possible types

Text content `string`

The contents of the developer message.

Array of content parts `array`

An array of content parts with a defined type. For developer messages, only type text is supported.

Show properties

text

string

required

The text content.

type

string

required

The type of content part. Always text.

role

string

required

The role of the messages author, in this case developer.

name

string

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

System message `object`

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

Show properties

content

string | array

required

The contents of the system message.

Show possible types

Text content `string`

The contents of the system message.

Array of content parts `array`

An array of content parts with a defined type. For system messages, only type text is supported.

Show possible types

text

string

required

The text content.

type

string

required

The type of content part. Always text.

role

string

required

The role of the messages author, in this case system.

name

string

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

Tool message `object`

Tool message containing the result of a tool call.

Show properties

content

string | array

required

The contents of the tool message.

Show possible types

Text content `string`

The contents of the tool message.

Array of content parts `array`

An array of content parts with a defined type. For tool messages, only type text is supported.

Show possible types

text

string

required

The text content.

type

string

required

The type of content part. Always text.

role

string

required

The role of the messages author, in this case tool.

tool_call_id

string

required

Tool call that this message is responding to.

User message `object`

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

Show properties

content

string | array

required

The contents of the user message.

Show possible types

Text content `string`

The text contents of the message.

Array of content parts `array`

An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs.

Show possible types

text

string

required

The text content.

type

string

required

The type of content part. Always text.

role

string

required

The role of the messages author, in this case user.

name

string

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

model

string

required

Available options:

gpt-oss-120b
zai-glm-4.7 (preview)

clear_thinking

boolean | null

Controls whether thinking content from previous conversation turns is included in the prompt context.Note: Thinking content from the current (latest unfinished) turn is always included regardless of this setting.

false - Thinking from all previous turns is preserved in the conversation history. Recommended for agentic workflows where reasoning from past tool-calling turns may be relevant for future tool calls.
true (default) - Thinking from earlier turns is excluded. Recommended for general chat conversations where reasoning from past turns is less relevant for performance.

When this parameter is not specified or set to null, the API defaults to clear_thinking: true.

This parameter is supported only on the zai-glm-4.7 model. For additional information, see Preserved thinking in the Z.ai documentation.

frequency_penalty

number | null

A number between -2.0 and 2.0. Positive values reduce the likelihood of the model repeating tokens by applying a penalty proportional to how frequently each token has already appeared in the generated output.Minimum: -2, Maximum: 2Default: 0

logit_bias

map | null

Modify the likelihood of specified tokens appearing in the completion. 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.Default: null

logprobs

bool

Whether to return log probabilities of the output tokens or not.Default: false

max_completion_tokens

integer | null

The maximum number of tokens that can be generated in the completion, including reasoning tokens. The total length of input tokens and generated tokens is limited by the model’s context length.

parallel_tool_calls

boolean | null

Whether to enable parallel function calling during tool use. When enabled (default), the model can request multiple tool calls simultaneously in a single response. When disabled, the model will only request one tool call at a time.Default: true

prediction

object | null

Configuration for a Predicted Output, which can greatly speed up response times when large parts of the model response are known in advance. This is most common when you are regenerating a file with mostly minor changes to the content.

Show possible types

Static Content

Static predicted output content, such as the content of a text or code file that is being regenerated.

Show properties

content

string | array

required

The content that should be matched when generating a model response. If continuous token sequences from the generated tokens match this content, the entire model response can be returned faster.

Show possible types

Text content `string`

The content used for a given Predicted Output. Typically the text of a file you are regenerating with only minor changes.

Array of content parts `array`

An array of content parts with a defined type. Supported options may differ based on the model that is used to generate the response. May contain text inputs.

Show possible types

text

string

required

The text content.

type

string

required

The type of content part. Always text.

type

string

required

The type of the predicted content you wish to provide. This type is currently always content.

Visit our page on Predicted Outputs for more information and examples.

presence_penalty

number | null

A number between -2.0 and 2.0. Positive values reduce the likelihood of the model repeating tokens that have already appeared in the output, encouraging the model to introduce new topics.Minimum: -2, Maximum: 2Default: 0

prompt_cache_key

string | null

An opaque identifier that groups related requests so they reuse the same prompt cache. Requests sharing the same prompt_cache_key are routed together, which increases cache hits and reduces time to first token.Set it to a stable identifier like a conversation ID, user ID, or session ID.Maximum length: 1024 charactersDefault: null

prompt_cache_key must be enabled on your account before you can use it. Contact us or reach out to your account representative to request access.

reasoning_effort

string | null

Controls the amount of reasoning the model performs. Supported values vary by model:gpt-oss-120b

"low" – Minimal reasoning, faster responses
"medium" – Moderate reasoning (default)
"high" – Extensive reasoning, more thorough analysis

zai-glm-4.7 (reasoning enabled by default)

"none" – Disables reasoning entirely

This parameter is only available for gpt-oss-120b and zai-glm-4.7 models.

response_format

object | null

An object that controls the format of the model response.Setting to { "type": "json_schema", "json_schema": { "name": "schema_name", "strict": true, "schema": {...} } } enforces schema compliance by ensuring that the model output conforms to your specified JSON schema. See Structured Outputs for more information.Setting { "type": "json_object" } enables the legacy JSON mode, ensuring that the model output is valid JSON. However, using json_schema is recommended for models that support it.

Show properties

Text `object`

Default response format. Generates plain text responses.

Show properties

type

string

required

The type of response format being defined. Always text.

JSON schema `object`

Generates structured JSON output that conforms to the specified schema. Use this format when you need the model to return structured JSON.

Show properties

json_schema

object

required

Structured Outputs configuration options.

Show properties

name

string

required

An optional name for your schema.

description

string

A description of the response format’s purpose, used by the model to determine how to generate its response in that format.

schema

object

A valid JSON Schema object that defines the structure, types, and requirements for the response. Supports standard JSON Schema features including types (string, number, boolean, integer, object, array, enum, anyOf, null), nested structures, required fields, and additionalProperties (must be set to false).

strict

boolean

When set to true, enforces strict adherence to the schema. The model will only return fields defined in the schema and with the correct types. When false, behaves similar to JSON mode but uses the schema as a guide. Defaults to false.

type

string

required

The type of response format being defined. Always json_schema.

JSON object `object`

A legacy method for generating JSON responses. Using json_schema is recommended for models that support it. To use json_object remember to also include a system or user message to specify the desired format.

Show properties

type

string

required

The type of response format being defined. Always json_object.

When using JSON object, you must explicitly instruct the model to generate JSON through a system or user message. json_object is not compatible with streaming - stream must be set to false.

seed

integer | null

If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result. Determinism is not guaranteed.

service_tier

string | null

Controls request prioritization.

This feature is in Private Preview. For access or more information, contact us or reach out to your account representative.

Available options:

priority - Highest priority processing (Only available for dedicated endpoints, not shared endpoints.)
default - Standard priority processing
auto - Automatically uses the highest available service tier
flex - Lowest priority processing

Default: defaultSee Service Tiers for more information.

stop

string | null

Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.

stream

boolean | null

If set, partial message deltas will be sent.

temperature

number | null

What sampling temperature to use, between 0 and 2.0. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both.Minimum: 0, Maximum: 2

tool_choice

string | object

Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {"type": "function", "function": {"name": "my_function"}} forces the model to call that tool.none is the default when no tools are present. auto is the default if tools are present.

tools

object | null

A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for.Specifying tools consumes prompt tokens in the context. If too many are given, the model may perform poorly or you may hit context length limitations

Show properties

tools.function.description

string

A description of what the function does, used by the model to choose when and how to call the function.

tools.function.name

string

The name of the function to be called.Supported characters: a-z, A-Z, 0-9, _, -Maximum length: 64Names that use other characters or exceed this length might work with some models, but compatibility isn’t guaranteed.

tools.function.parameters

object

The parameters the functions accepts, described as a JSON Schema object. Omitting parameters defines a function with an empty parameter list.

tools.type

string

The type of the tool. Currently, only function is supported.

top_logprobs

integer | null

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. logprobs must be set to true if this parameter is used.Minimum: 0, Maximum: 20

top_p

number | null

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So, 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both.Minimum: 0, Maximum: 1

user

string | null

A unique identifier representing your end-user, which can help to monitor and detect abuse.

Response

string

A unique identifier for the chat completion.

choices

object[]

A list of chat completion choices. Can be more than one if n is greater than 1.

Show choice properties

finish_reason

string

The reason the model stopped generating tokens. Possible values: stop, length, content_filter, tool_calls.

index

integer

The index of the choice in the list of choices.

logprobs

object | null

Log probability information for the output tokens.

reasoning_logprobs

object | null

Log probability information for the reasoning tokens, if provided by the model.

message

object

A chat completion message generated by the model.

Show message properties

content

string

The contents of the message.

role

string

The role of the author of this message.

reasoning

string

The model’s reasoning content when using reasoning models.

tool_calls

array | null

A list of tool calls the model requested in its response.

Show possible types

Tool call `object`

string

The ID of the tool call.

type

string

The type of tool call, currently only function is supported.

function

object

The function call details.

Show properties

name

string

The name of the function to call.

arguments

string

The JSON-encoded arguments supplied by the model.

created

integer

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

model

string

The model used for the chat completion.

object

string

The object type, which is always chat.completion.

system_fingerprint

string

A fingerprint for the model or backend used to generate the response.

service_tier

string

The service tier used for the request, or null if not specified.

service_tier_used

string

The service tier used for processing the request. Only present when service_tier is set to auto in the request.Possible values: priority, default, flex

usage

object

Usage statistics for the completion request.

Show usage properties

prompt_tokens

integer

Number of tokens in the prompt.

completion_tokens

integer

Number of tokens in the generated completion.

total_tokens

integer

Total number of tokens used in the request (prompt + completion).

prompt_tokens_details

object

Detailed breakdown of prompt token usage.

Show properties

cached_tokens

integer

Number of prompt tokens that were served from the cache and reused from a previous request. See Prompt Caching for more information.

completion_tokens_details

object

Breakdown of completion tokens when using Predicted Outputs.

Show properties

accepted_prediction_tokens

integer

When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion.

rejected_prediction_tokens

integer

When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. Like reasoning tokens, these tokens are still counted in the total completion tokens for the purposes of billing, output, and context window limits.

reasoning_tokens

integer

Tokens spent on model reasoning, if applicable.

time_info

object

Performance timing information for the request.

Show time_info properties

queue_time

number

Time spent in queue waiting for processing (in seconds).

prompt_time

number

Time spent processing the prompt/input tokens (in seconds).

completion_time

number

Time spent generating the completion/output tokens (in seconds).

total_time

number

Total time for the entire request from submission to completion (in seconds).

created

number

Unix timestamp (in seconds) of when the time_info was recorded.

from cerebras.cloud.sdk import Cerebras
import os 

client = Cerebras(api_key=os.environ.get("CEREBRAS_API_KEY"),)

chat_completion = client.chat.completions.create(
    model="gpt-oss-120b",
    messages=[
        {"role": "user", "content": "Hello!",}
    ],
)
print(chat_completion)

{
  "id": "chatcmpl-30b3c3d8-ca41-48e7-9ef0-27e322604a13",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! 👋 How can I help you today?",
        "reasoning": "The user just says \"Hello!\" with no further context.\n\nWe need to respond politely, maybe ask how can assist.\n\nWe can also be friendly.\n\nNo constraints. We'll respond as chat.",
        "tool_calls": null
      },
      "logprobs": null,
      "reasoning_logprobs": null
    }
  ],
  "created": 1769729480,
  "model": "gpt-oss-120b",
  "object": "chat.completion",
  "system_fingerprint": "fp_e7ab83753cbd28777b40",
  "time_info": {
    "completion_time": 0.04040406,
    "prompt_time": 0.00383762,
    "queue_time": 0.004628115,
    "total_time": 0.05013155937194824,
    "created": 1769729480.0787008
  },
  "usage": {
    "completion_tokens": 59,
    "completion_tokens_details": {
      "accepted_prediction_tokens": 0,
      "rejected_prediction_tokens": 0,
      "reasoning_tokens": 0
    },
    "prompt_tokens": 69,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "total_tokens": 128
  },
  "service_tier": null
}

Versions

Completions

⌘I

Documentation Index

​Request

​Headers

​Body

Assistant message object

Text content string

Array of content parts array

Developer message object

Text content string

Array of content parts array

System message object

Text content string

Array of content parts array

Tool message object

Text content string

Array of content parts array

User message object

Text content string

Array of content parts array

Static Content

Text content string

Array of content parts array

Text object

JSON schema object

JSON object object

​Response

Tool call object

Request

Headers

Body

Assistant message `object`

Text content `string`

Array of content parts `array`

Developer message `object`

Text content `string`

Array of content parts `array`

System message `object`

Text content `string`

Array of content parts `array`

Tool message `object`

Text content `string`

Array of content parts `array`

User message `object`

Text content `string`

Array of content parts `array`

Text content `string`

Array of content parts `array`

Text `object`

JSON schema `object`

JSON object `object`

Response

Tool call `object`