openrouter-python-sdk-retry-errors/docs/sdks/chat

Chat

(chat)

Overview

Chat completion operations

Available Operations

• complete - Create a chat completion
• complete_stream - Create a chat completion

complete

Creates a model response for the given chat conversation. Supports both streaming and non-streaming modes.

Example Usage

from openrouter import OpenRouter, models
import os


with OpenRouter(
    bearer_auth=os.getenv("OPENROUTER_BEARER_AUTH", ""),
) as open_router:

    res = open_router.chat.complete(messages=[
        {
            "role": models.ChatCompletionUserMessageParamRole.USER,
            "content": "Hello, how are you?",
        },
    ], stream=False, temperature=1, top_p=1)

    # Handle response
    print(res)

Parameters

Parameter	Type	Required	Description	Example
messages	List[models.ChatCompletionMessageParam]	✔️	List of messages for the conversation	[ { "role": "user", "content": "Hello, how are you?" } ]
model	Optional[str]	➖	Model to use for completion
frequency_penalty	OptionalNullable[float]	➖	Frequency penalty (-2.0 to 2.0)
logit_bias	Dict[str, float]	➖	Token logit bias adjustments
logprobs	OptionalNullable[bool]	➖	Return log probabilities
top_logprobs	OptionalNullable[float]	➖	Number of top log probabilities to return (0-20)
max_completion_tokens	OptionalNullable[float]	➖	Maximum tokens in completion
max_tokens	OptionalNullable[float]	➖	Maximum tokens (deprecated, use max_completion_tokens)
metadata	Dict[str, str]	➖	Key-value pairs for additional object information (max 16 pairs, 64 char keys, 512 char values)
presence_penalty	OptionalNullable[float]	➖	Presence penalty (-2.0 to 2.0)
reasoning	OptionalNullable[models.ChatCompletionCreateParamsReasoning]	➖	Reasoning configuration
response_format	Optional[models.ChatCompletionCreateParamsResponseFormatUnion]	➖	Response format configuration
seed	OptionalNullable[int]	➖	Random seed for deterministic outputs
stop	OptionalNullable[models.ChatCompletionCreateParamsStop]	➖	Stop sequences (up to 4)
stream	OptionalNullable[bool]	➖	Enable streaming response
stream_options	OptionalNullable[models.ChatCompletionCreateParamsStreamOptions]	➖	N/A
temperature	OptionalNullable[float]	➖	Sampling temperature (0-2)
tool_choice	Optional[models.ChatCompletionToolChoiceOption]	➖	Tool choice configuration
tools	List[models.ChatCompletionTool]	➖	Available tools for function calling
top_p	OptionalNullable[float]	➖	Nucleus sampling parameter (0-1)
user	Optional[str]	➖	Unique user identifier
models_llm	List[str]	➖	Order of models to fallback to for this request
reasoning_effort	OptionalNullable[models.ChatCompletionCreateParamsReasoningEffort]	➖	Reasoning effort
provider	OptionalNullable[models.ChatCompletionCreateParamsProvider]	➖	When multiple model providers are available, optionally indicate your routing preference.
plugins	List[models.ChatCompletionCreateParamsPluginUnion]	➖	Plugins you want to enable for this request, including their settings.
retries	Optional[utils.RetryConfig]	➖	Configuration to override the default retry behavior of the client.

Response

models.ChatCompletion

Errors

Error Type	Status Code	Content Type
errors.ChatCompletionError	400, 401, 429	application/json
errors.ChatCompletionError	500	application/json
errors.OpenRouterDefaultError	4XX, 5XX	*/*

complete_stream

Creates a model response for the given chat conversation. Supports both streaming and non-streaming modes.

Example Usage

from openrouter import OpenRouter, models
import os


with OpenRouter(
    bearer_auth=os.getenv("OPENROUTER_BEARER_AUTH", ""),
) as open_router:

    res = open_router.chat.complete_stream(messages=[
        {
            "role": models.ChatCompletionUserMessageParamRole.USER,
            "content": "Hello, how are you?",
        },
    ], stream=True, temperature=1, top_p=1)

    with res as event_stream:
        for event in event_stream:
            # handle event
            print(event, flush=True)

Parameters

Parameter	Type	Required	Description	Example
messages	List[models.ChatCompletionMessageParam]	✔️	List of messages for the conversation	[ { "role": "user", "content": "Hello, how are you?" } ]
model	Optional[str]	➖	Model to use for completion
frequency_penalty	OptionalNullable[float]	➖	Frequency penalty (-2.0 to 2.0)
logit_bias	Dict[str, float]	➖	Token logit bias adjustments
logprobs	OptionalNullable[bool]	➖	Return log probabilities
top_logprobs	OptionalNullable[float]	➖	Number of top log probabilities to return (0-20)
max_completion_tokens	OptionalNullable[float]	➖	Maximum tokens in completion
max_tokens	OptionalNullable[float]	➖	Maximum tokens (deprecated, use max_completion_tokens)
metadata	Dict[str, str]	➖	Key-value pairs for additional object information (max 16 pairs, 64 char keys, 512 char values)
presence_penalty	OptionalNullable[float]	➖	Presence penalty (-2.0 to 2.0)
reasoning	OptionalNullable[models.ChatStreamCompletionCreateParamsReasoning]	➖	Reasoning configuration
response_format	Optional[models.ChatStreamCompletionCreateParamsResponseFormatUnion]	➖	Response format configuration
seed	OptionalNullable[int]	➖	Random seed for deterministic outputs
stop	OptionalNullable[models.ChatStreamCompletionCreateParamsStop]	➖	Stop sequences (up to 4)
stream	Optional[bool]	➖	Enable streaming response
stream_options	OptionalNullable[models.ChatStreamCompletionCreateParamsStreamOptions]	➖	N/A
temperature	OptionalNullable[float]	➖	Sampling temperature (0-2)
tool_choice	Optional[models.ChatCompletionToolChoiceOption]	➖	Tool choice configuration
tools	List[models.ChatCompletionTool]	➖	Available tools for function calling
top_p	OptionalNullable[float]	➖	Nucleus sampling parameter (0-1)
user	Optional[str]	➖	Unique user identifier
models_llm	List[str]	➖	Order of models to fallback to for this request
reasoning_effort	OptionalNullable[models.ChatStreamCompletionCreateParamsReasoningEffort]	➖	Reasoning effort
provider	OptionalNullable[models.ChatStreamCompletionCreateParamsProvider]	➖	When multiple model providers are available, optionally indicate your routing preference.
plugins	List[models.ChatStreamCompletionCreateParamsPluginUnion]	➖	Plugins you want to enable for this request, including their settings.
retries	Optional[utils.RetryConfig]	➖	Configuration to override the default retry behavior of the client.

Response

Union[eventstreaming.EventStream[models.StreamChatCompletionResponseBody], eventstreaming.EventStreamAsync[models.StreamChatCompletionResponseBody]]

Errors

Error Type	Status Code	Content Type
errors.ChatCompletionError	400, 401, 429	application/json
errors.ChatCompletionError	500	application/json
errors.OpenRouterDefaultError	4XX, 5XX	*/*