Module openrouter

ballerinax/openrouter Ballerina library

1.0.0

Overview

OpenRouter is a unified API platform that provides access to multiple AI language models from various providers, allowing developers to easily switch between and compare different models through a single interface.

The ballerinax/openrouter package offers APIs to connect and interact with OpenRouter API endpoints, specifically based on OpenRouter API v1.

Setup guide

To use the OpenRouter connector, you must have access to the OpenRouter API through an OpenRouter developer account and obtain an API access token. If you do not have an OpenRouter account, you can sign up for one here.

Step 1: Create an OpenRouter Account

Navigate to the OpenRouter website and sign up for an account or log in if you already have one.
Note that OpenRouter operates on a pay-per-use model with credits, and API access is available to all registered users regardless of plan type.

Step 2: Generate an API Access Token

Log in to your OpenRouter account.
Navigate to the Keys section by clicking on your profile in the top right corner and selecting "Keys" from the dropdown menu.
Click on "Create Key" to generate a new API key.
Provide a name for your key and set any optional restrictions if needed, then click "Create Key".

Tip: You must copy and store this key somewhere safe. It won't be visible again in your account settings for security reasons.

Quickstart

To use the openrouter connector in your Ballerina application, update the .bal file as follows:

Step 1: Import the module


import ballerinax/openrouter;

Step 2: Instantiate a new connector

Create a Config.toml file and configure the obtained access token:


token = "<Your_OpenRouter_API_Token>"

Create an openrouter:ConnectionConfig and initialize the client:


configurable string token = ?;

final openrouter:Client openrouterClient = check new({
    auth: {
        token
    }
});

Step 3: Invoke the connector operation

Now, utilize the available connector operations.

Create a chat completion


public function main() returns error? {
    openrouter:ChatGenerationParams chatRequest = {
        model: "openai/gpt-3.5-turbo",
        messages: [
            {
                role: "user",
                content: "Hello! Can you help me write a simple greeting message?"
            }
        ],
        max_tokens: 150,
        temperature: 0.7
    };

    openrouter:ChatResponse response = check openrouterClient->/chat/completions.post(chatRequest);
}

Step 4: Run the Ballerina application


bal run

Examples

The openrouter connector provides practical examples illustrating usage in various scenarios. Explore these examples, covering the following use cases:

Ai model deployment pipeline - Demonstrates how to automate AI model deployment processes using Ballerina connector for OpenRouter.
Ai security audit - Illustrates conducting security audits for AI systems and models.

Clients

openrouter: Client

Isolated

OpenAI-compatible API with additional OpenRouter features

Constructor

Gets invoked to initialize the connector.

init (ConnectionConfig config, string serviceUrl)

config ConnectionConfig - The configurations to be used when initializing the connector

serviceUrl string "https://openrouter.ai/api/v1" - URL of the target service

post responses

Isolated FunctionResource Function

function post responses(OpenResponsesRequest payload, CreateResponsesHeaders headers) returns OpenResponsesNonStreamingResponse|error

Create a response

Parameters

payload OpenResponsesRequest -

headers CreateResponsesHeaders (default {}) - Headers to be sent with the request

Return Type

OpenResponsesNonStreamingResponse|error - Successful response

post messages

Isolated FunctionResource Function

function post messages(AnthropicMessagesRequest payload, CreateMessagesHeaders headers) returns AnthropicMessagesResponse|error

Create a message

Parameters

payload AnthropicMessagesRequest -

headers CreateMessagesHeaders (default {}) - Headers to be sent with the request

Return Type

AnthropicMessagesResponse|error - Successful response

get activity

Isolated FunctionResource Function

function get activity(GetUserActivityHeaders headers, *GetUserActivityQueries queries) returns ActivityResponse|error

Get user activity grouped by endpoint

Parameters

headers GetUserActivityHeaders (default {}) - Headers to be sent with the request

queries *GetUserActivityQueries - Queries to be sent with the request

Return Type

ActivityResponse|error - Returns user activity data grouped by endpoint

post chat/completions

Isolated FunctionResource Function

function post chat/completions(ChatGenerationParams payload, SendChatCompletionRequestHeaders headers) returns ChatResponse|error

Create a chat completion

Parameters

payload ChatGenerationParams -

headers SendChatCompletionRequestHeaders (default {}) - Headers to be sent with the request

Return Type

ChatResponse|error - Successful chat completion response

get credits

Isolated FunctionResource Function

function get credits(GetCreditsHeaders headers) returns CreditsResponse|error

Get remaining credits

Parameters

headers GetCreditsHeaders (default {}) - Headers to be sent with the request

Return Type

CreditsResponse|error - Returns the total credits purchased and used

post credits/coinbase

Isolated FunctionResource Function

function post credits/coinbase(CreateChargeRequest payload, CreateCoinbaseChargeHeaders headers) returns CoinbaseCreditResponse|error

Create a Coinbase charge for crypto payment

Parameters

payload CreateChargeRequest -

headers CreateCoinbaseChargeHeaders (default {}) - Headers to be sent with the request

Return Type

CoinbaseCreditResponse|error - Returns the calldata to fulfill the transaction

post embeddings

Isolated FunctionResource Function

function post embeddings(CreateEmbeddingsRequest payload, CreateEmbeddingsHeaders headers) returns CreateEmbeddingsResponse|error

Submit an embedding request

Parameters

payload CreateEmbeddingsRequest -

headers CreateEmbeddingsHeaders (default {}) - Headers to be sent with the request

Return Type

CreateEmbeddingsResponse|error - Embedding response

get embeddings/models

Isolated FunctionResource Function

function get embeddings/models(ListEmbeddingsModelsHeaders headers) returns ModelsListResponse|error

List all embeddings models

Parameters

headers ListEmbeddingsModelsHeaders (default {}) - Headers to be sent with the request

Return Type

ModelsListResponse|error - Returns a list of embeddings models

get generation

Isolated FunctionResource Function

function get generation(GetGenerationHeaders headers, *GetGenerationQueries queries) returns GenerationResponse|error

Get request & usage metadata for a generation

Parameters

headers GetGenerationHeaders (default {}) - Headers to be sent with the request

queries *GetGenerationQueries - Queries to be sent with the request

Return Type

GenerationResponse|error - Returns the request metadata for this generation

get models/count

Isolated FunctionResource Function

function get models/count(ListModelsCountHeaders headers) returns ModelsCountResponse|error

Get total count of available models

Parameters

headers ListModelsCountHeaders (default {}) - Headers to be sent with the request

Return Type

ModelsCountResponse|error - Returns the total count of available models

get models

Isolated FunctionResource Function

function get models(GetModelsHeaders headers, *GetModelsQueries queries) returns ModelsListResponse|error

List all models and their properties

Parameters

headers GetModelsHeaders (default {}) - Headers to be sent with the request

queries *GetModelsQueries - Queries to be sent with the request

Return Type

ModelsListResponse|error - Returns a list of models or RSS feed

get models/user

Isolated FunctionResource Function

function get models/user(ListModelsUserHeaders headers) returns ModelsListResponse|error

List models filtered by user provider preferences, privacy settings, and guardrails

Parameters

headers ListModelsUserHeaders (default {}) - Headers to be sent with the request

Return Type

ModelsListResponse|error - Returns a list of models filtered by user provider preferences

get models/[string author]/[string slug]/endpoints

Isolated FunctionResource Function

function get models/[string author]/[string slug]/endpoints(ListEndpointsHeaders headers) returns ModelEndpointsResponse|error

List all endpoints for a model

Parameters

headers ListEndpointsHeaders (default {}) - Headers to be sent with the request

Return Type

ModelEndpointsResponse|error - Returns a list of endpoints

get endpoints/zdr

Isolated FunctionResource Function

function get endpoints/zdr(ListEndpointsZdrHeaders headers) returns ZdrEndpointsResponse|error

Preview the impact of ZDR on the available endpoints

Parameters

headers ListEndpointsZdrHeaders (default {}) - Headers to be sent with the request

Return Type

ZdrEndpointsResponse|error - Returns a list of endpoints

get providers

Isolated FunctionResource Function

function get providers(ListProvidersHeaders headers) returns ProvidersResponse|error

List all providers

Parameters

headers ListProvidersHeaders (default {}) - Headers to be sent with the request

Return Type

ProvidersResponse|error - Returns a list of providers

get keys

Isolated FunctionResource Function

function get keys(ListHeaders headers, *ListQueries queries) returns KeysListResponse|error

List API keys

Parameters

headers ListHeaders (default {}) - Headers to be sent with the request

queries *ListQueries - Queries to be sent with the request

Return Type

KeysListResponse|error - List of API keys

post keys

Isolated FunctionResource Function

function post keys(CreateKeyRequest payload, CreateKeysHeaders headers) returns CreateKeyResponse|error

Create a new API key

Parameters

payload CreateKeyRequest -

headers CreateKeysHeaders (default {}) - Headers to be sent with the request

Return Type

CreateKeyResponse|error - API key created successfully

get keys/[string hash]

Isolated FunctionResource Function

function get keys/[string hash](GetKeyHeaders headers) returns KeyDetailsResponse|error

Get a single API key

Parameters

headers GetKeyHeaders (default {}) - Headers to be sent with the request

Return Type

KeyDetailsResponse|error - API key details

delete keys/[string hash]

Isolated FunctionResource Function

function delete keys/[string hash](DeleteKeysHeaders headers) returns DeleteKeyResponse|error

Delete an API key

Parameters

headers DeleteKeysHeaders (default {}) - Headers to be sent with the request

Return Type

DeleteKeyResponse|error - API key deleted successfully

patch keys/[string hash]

Isolated FunctionResource Function

function patch keys/[string hash](UpdateKeyRequest payload, UpdateKeysHeaders headers) returns UpdateKeyResponse|error

Update an API key

Parameters

payload UpdateKeyRequest -

headers UpdateKeysHeaders (default {}) - Headers to be sent with the request

Return Type

UpdateKeyResponse|error - API key updated successfully

get guardrails

Isolated FunctionResource Function

function get guardrails(ListGuardrailsHeaders headers, *ListGuardrailsQueries queries) returns GuardrailListResponse|error

List guardrails

Parameters

headers ListGuardrailsHeaders (default {}) - Headers to be sent with the request

queries *ListGuardrailsQueries - Queries to be sent with the request

Return Type

GuardrailListResponse|error - List of guardrails

post guardrails

Isolated FunctionResource Function

function post guardrails(CreateGuardrailRequest payload, CreateGuardrailHeaders headers) returns CreateGuardrailResponse|error

Create a guardrail

Parameters

payload CreateGuardrailRequest -

headers CreateGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

CreateGuardrailResponse|error - Guardrail created successfully

get guardrails/[string id]

Isolated FunctionResource Function

function get guardrails/[string id](GetGuardrailHeaders headers) returns GetGuardrailResponse|error

Get a guardrail

Parameters

headers GetGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

GetGuardrailResponse|error - Guardrail details

delete guardrails/[string id]

Isolated FunctionResource Function

function delete guardrails/[string id](DeleteGuardrailHeaders headers) returns DeleteGuardrailResponse|error

Delete a guardrail

Parameters

headers DeleteGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

DeleteGuardrailResponse|error - Guardrail deleted successfully

patch guardrails/[string id]

Isolated FunctionResource Function

function patch guardrails/[string id](UpdateGuardrailRequest payload, UpdateGuardrailHeaders headers) returns UpdateGuardrailResponse|error

Update a guardrail

Parameters

payload UpdateGuardrailRequest -

headers UpdateGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

UpdateGuardrailResponse|error - Guardrail updated successfully

get guardrails/assignments/keys

Isolated FunctionResource Function

function get guardrails/assignments/keys(ListKeyAssignmentsHeaders headers, *ListKeyAssignmentsQueries queries) returns GuardrailKeyAssignmentsResponse|error

List all key assignments

Parameters

headers ListKeyAssignmentsHeaders (default {}) - Headers to be sent with the request

queries *ListKeyAssignmentsQueries - Queries to be sent with the request

Return Type

GuardrailKeyAssignmentsResponse|error - List of key assignments

get guardrails/assignments/members

Isolated FunctionResource Function

function get guardrails/assignments/members(ListMemberAssignmentsHeaders headers, *ListMemberAssignmentsQueries queries) returns GuardrailMemberAssignmentsResponse|error

List all member assignments

Parameters

headers ListMemberAssignmentsHeaders (default {}) - Headers to be sent with the request

queries *ListMemberAssignmentsQueries - Queries to be sent with the request

Return Type

GuardrailMemberAssignmentsResponse|error - List of member assignments

get guardrails/[string id]/assignments/keys

Isolated FunctionResource Function

function get guardrails/[string id]/assignments/keys(ListGuardrailKeyAssignmentsHeaders headers, *ListGuardrailKeyAssignmentsQueries queries) returns GuardrailKeyAssignmentsResponse|error

List key assignments for a guardrail

Parameters

headers ListGuardrailKeyAssignmentsHeaders (default {}) - Headers to be sent with the request

queries *ListGuardrailKeyAssignmentsQueries - Queries to be sent with the request

Return Type

GuardrailKeyAssignmentsResponse|error - List of key assignments

post guardrails/[string id]/assignments/keys

Isolated FunctionResource Function

function post guardrails/[string id]/assignments/keys(AssignKeysRequest payload, BulkAssignKeysToGuardrailHeaders headers) returns AssignKeysResponse|error

Bulk assign keys to a guardrail

Parameters

payload AssignKeysRequest -

headers BulkAssignKeysToGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

AssignKeysResponse|error - Assignment result

get guardrails/[string id]/assignments/members

Isolated FunctionResource Function

function get guardrails/[string id]/assignments/members(ListGuardrailMemberAssignmentsHeaders headers, *ListGuardrailMemberAssignmentsQueries queries) returns GuardrailMemberAssignmentsResponse|error

List member assignments for a guardrail

Parameters

headers ListGuardrailMemberAssignmentsHeaders (default {}) - Headers to be sent with the request

queries *ListGuardrailMemberAssignmentsQueries - Queries to be sent with the request

Return Type

GuardrailMemberAssignmentsResponse|error - List of member assignments

post guardrails/[string id]/assignments/members

Isolated FunctionResource Function

function post guardrails/[string id]/assignments/members(AssignMembersRequest payload, BulkAssignMembersToGuardrailHeaders headers) returns AssignMembersResponse|error

Bulk assign members to a guardrail

Parameters

payload AssignMembersRequest -

headers BulkAssignMembersToGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

AssignMembersResponse|error - Assignment result

post guardrails/[string id]/assignments/keys/remove

Isolated FunctionResource Function

function post guardrails/[string id]/assignments/keys/remove(RemoveKeysRequest payload, BulkUnassignKeysFromGuardrailHeaders headers) returns RemoveKeysResponse|error

Bulk unassign keys from a guardrail

Parameters

payload RemoveKeysRequest -

headers BulkUnassignKeysFromGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

RemoveKeysResponse|error - Unassignment result

post guardrails/[string id]/assignments/members/remove

Isolated FunctionResource Function

function post guardrails/[string id]/assignments/members/remove(RemoveMembersRequest payload, BulkUnassignMembersFromGuardrailHeaders headers) returns RemoveMembersResponse|error

Bulk unassign members from a guardrail

Parameters

payload RemoveMembersRequest -

headers BulkUnassignMembersFromGuardrailHeaders (default {}) - Headers to be sent with the request

Return Type

RemoveMembersResponse|error - Unassignment result

get 'key

Isolated FunctionResource Function

function get 'key(GetCurrentKeyHeaders headers) returns CurrentKeyResponse|error

Get current API key

Parameters

headers GetCurrentKeyHeaders (default {}) - Headers to be sent with the request

Return Type

CurrentKeyResponse|error - API key details

post auth/keys

Isolated FunctionResource Function

function post auth/keys(AuthKeyRequest payload, ExchangeAuthCodeForAPIKeyHeaders headers) returns AuthKeyResponse|error

Exchange authorization code for API key

Parameters

payload AuthKeyRequest -

headers ExchangeAuthCodeForAPIKeyHeaders (default {}) - Headers to be sent with the request

Return Type

AuthKeyResponse|error - Successfully exchanged code for an API key

post auth/keys/code

Isolated FunctionResource Function

function post auth/keys/code(KeyCodeRequest payload, CreateAuthKeysCodeHeaders headers) returns KeyCodeResponse|error

Create authorization code

Parameters

payload KeyCodeRequest -

headers CreateAuthKeysCodeHeaders (default {}) - Headers to be sent with the request

Return Type

KeyCodeResponse|error - Successfully created authorization code

Records

openrouter: ActivityItem

Fields

date string - Date of the activity (YYYY-MM-DD format)

model string - Model slug (e.g., "openai/gpt-4.1")

model_permaslug string - Model permaslug (e.g., "openai/gpt-4.1-2025-04-14")

endpoint_id string - Unique identifier for the endpoint

provider_name string - Name of the provider serving this endpoint

usage decimal - Total cost in USD (OpenRouter credits spent)

byok_usage_inference decimal - BYOK inference cost in USD (external credits spent)

requests decimal - Number of requests made

prompt_tokens decimal - Total prompt tokens used

completion_tokens decimal - Total completion tokens generated

reasoning_tokens decimal - Total reasoning tokens used

openrouter: ActivityResponse

Fields

data ActivityItem[] - List of activity items

openrouter: AnthropicCacheControl

Fields

'type "ephemeral" -

ttl? "5m"|"1h" -

openrouter: AnthropicMessagesMetadata

Fields

user_id? string -

openrouter: AnthropicMessagesProvider

Closed record

When multiple model providers are available, optionally indicate your routing preference.

Fields

allow_fallbacks? boolean - Whether to allow backup providers to serve requests
- true: (default) when the primary provider (or your custom providers in "order") is unavailable, use the next best provider.
- false: use only the primary/custom provider, and return the upstream error if it's unavailable.

require_parameters? boolean - Whether to filter providers to only those that support the parameters you've provided. If this setting is omitted or set to false, then providers will receive only the parameters they support, and ignore the rest.

data_collection? DataCollection - Data collection setting. If no available model provider meets the requirement, your request will return an error.
- allow: (default) allow providers which store user data non-transiently and may train on it
- deny: use only providers which do not collect user data.

zdr? boolean - Whether to restrict routing to only ZDR (Zero Data Retention) endpoints. When true, only endpoints that do not retain prompts will be used.

enforce_distillable_text? boolean - Whether to restrict routing to only models that allow text distillation. When true, only models where the author has allowed distillation will be used.

'order? (ProviderName|string)[] - An ordered list of provider slugs. The router will attempt to use the first provider in the subset of this list that supports your requested model, and fall back to the next if it is unavailable. If no providers are available, the request will fail with an error message.

only? (ProviderName|string)[] - List of provider slugs to allow. If provided, this list is merged with your account-wide allowed provider settings for this request.

ignore? (ProviderName|string)[] - List of provider slugs to ignore. If provided, this list is merged with your account-wide ignored provider settings for this request.

quantizations? Quantization[] - A list of quantization levels to filter the provider by.

sort? ProviderSort -

max_price? OpenResponsesRequest_provider_max_price - The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion.

preferred_min_throughput? PreferredMinThroughput - Preferred minimum throughput (in tokens per second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints below the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

preferred_max_latency? PreferredMaxLatency - Preferred maximum latency (in seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints above the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

openrouter: AnthropicMessagesRequest

Request schema for Anthropic Messages API endpoint

Fields

model string -

max_tokens decimal -

messages OpenRouterAnthropicMessageParam[] -

system? string|record { 'type "text" , text string, citations (record { 'type "char_location" , cited_text string, document_index decimal, document_title string, start_char_index decimal, end_char_index decimal }|record { 'type "page_location" , cited_text string, document_index decimal, document_title string, start_page_number decimal, end_page_number decimal }|record { 'type "content_block_location" , cited_text string, document_index decimal, document_title string, start_block_index decimal, end_block_index decimal }|record { 'type "web_search_result_location" , cited_text string, encrypted_index string, title string, url string }|record { 'type "search_result_location" , cited_text string, search_result_index decimal, 'source string, title string, start_block_index decimal, end_block_index decimal })[], cache_control record { 'type "ephemeral" , ttl "5m"|"1h" } }[] -

metadata? AnthropicMessagesMetadata -

stop_sequences? string[] -

'stream? boolean -

temperature? decimal -

top_p? decimal -

top_k? decimal -

tools? (anydata)[] -

tool_choice? record { 'type "auto" , disable_parallel_tool_use boolean }|record { 'type "any" , disable_parallel_tool_use boolean }|record { 'type "none" }|record { 'type "tool" , name string, disable_parallel_tool_use boolean } -

thinking? record { 'type "enabled" , budget_tokens decimal }|record { 'type "disabled" }|record { 'type "adaptive" } -

service_tier? "auto"|"standard_only" -

cache_control? AnthropicCacheControl -

provider? AnthropicMessagesProvider - When multiple model providers are available, optionally indicate your routing preference.

plugins? (record { id "auto-router" , enabled boolean, allowed_models string[] }|record { id "moderation" }|record { id "web" , enabled boolean, max_results decimal, search_prompt string, engine WebSearchEngine, native_tool_version "web_search_20250305"|"web_search_20260209" , include_domains string[], exclude_domains string[] }|record { id "file-parser" , enabled boolean, pdf PDFParserOptions }|record { id "response-healing" , enabled boolean })[] - Plugins you want to enable for this request, including their settings.

route? "fallback"|"sort" - DEPRECATED Use providers.sort.partition instead. Backwards-compatible alias for providers.sort.partition. Accepts legacy values: "fallback" (maps to "model"), "sort" (maps to "none").

user? string - A unique identifier representing your end-user, which helps distinguish between different users of your app. This allows your app to identify specific users in case of abuse reports, preventing your entire app from being affected by the actions of individual users. Maximum of 128 characters.

session_id? string - A unique identifier for grouping related requests (e.g., a conversation or agent workflow) for observability. If provided in both the request body and the x-session-id header, the body value takes precedence. Maximum of 128 characters.

trace? OpenResponsesRequest_trace - Metadata for observability and tracing. Known keys (trace_id, trace_name, span_name, generation_name, parent_span_id) have special handling. Additional keys are passed through as custom metadata to configured broadcast destinations.

models? string[] -

output_config? AnthropicOutputConfig - Configuration for controlling output behavior. Currently supports the effort parameter for Claude Opus 4.5.

openrouter: AnthropicMessagesResponse

Non-streaming response from the Anthropic Messages API with OpenRouter extensions

Fields

Fields Included from *BaseAnthropicMessagesResponse

id string
type "message"
role "assistant"
content anydata[]
model string
stop_reason "end_turn"|"max_tokens"|"stop_sequence"|"tool_use"|"pause_turn"|"refusal"
stop_sequence string
usage BaseAnthropicMessagesResponse_usage
anydata...

provider? "AnyScale"|"Atoma"|"Cent-ML"|"CrofAI"|"Enfer"|"GoPomelo"|"HuggingFace"|"Hyperbolic 2"|"InoCloud"|"Kluster"|"Lambda"|"Lepton"|"Lynn 2"|"Lynn"|"Mancer"|"Meta"|"Modal"|"Nineteen"|"OctoAI"|"Recursal"|"Reflection"|"Replicate"|"SambaNova 2"|"SF Compute"|"Targon"|"Together 2"|"Ubicloud"|"01.AI"|"AI21"|"AionLabs"|"Alibaba"|"Ambient"|"Amazon Bedrock"|"Amazon Nova"|"Anthropic"|"Arcee AI"|"AtlasCloud"|"Avian"|"Azure"|"BaseTen"|"BytePlus"|"Black Forest Labs"|"Cerebras"|"Chutes"|"Cirrascale"|"Clarifai"|"Cloudflare"|"Cohere"|"Crusoe"|"DeepInfra"|"DeepSeek"|"Featherless"|"Fireworks"|"Friendli"|"GMICloud"|"Google"|"Google AI Studio"|"Groq"|"Hyperbolic"|"Inception"|"Inceptron"|"InferenceNet"|"Ionstream"|"Infermatic"|"Io Net"|"Inflection"|"Liquid"|"Mara"|"Mancer 2"|"Minimax"|"ModelRun"|"Mistral"|"Modular"|"Moonshot AI"|"Morph"|"NCompass"|"Nebius"|"NextBit"|"Novita"|"Nvidia"|"OpenAI"|"OpenInference"|"Parasail"|"Perplexity"|"Phala"|"Relace"|"SambaNova"|"Seed"|"SiliconFlow"|"Sourceful"|"StepFun"|"Stealth"|"StreamLake"|"Switchpoint"|"Together"|"Upstage"|"Venice"|"WandB"|"Xiaomi"|"xAI"|"Z.AI"|"FakeProvider" -

openrouter: AnthropicOutputConfig

Configuration for controlling output behavior. Currently supports the effort parameter for Claude Opus 4.5.

Fields

effort? "low"|"medium"|"high"|"max" - How much effort the model should put into its response. Higher effort levels may result in more thorough analysis but take longer. Valid values are low, medium, high, or max.

openrouter: ApiKeyInfo

The API key information

Fields

hash string - Unique hash identifier for the API key

name string - Name of the API key

label string - Human-readable label for the API key

disabled boolean - Whether the API key is disabled

'limit decimal - Spending limit for the API key in USD

limit_remaining decimal - Remaining spending limit in USD

limit_reset string - Type of limit reset for the API key

include_byok_in_limit boolean - Whether to include external BYOK usage in the credit limit

usage decimal - Total OpenRouter credit usage (in USD) for the API key

usage_daily decimal - OpenRouter credit usage (in USD) for the current UTC day

usage_weekly decimal - OpenRouter credit usage (in USD) for the current UTC week (Monday-Sunday)

usage_monthly decimal - OpenRouter credit usage (in USD) for the current UTC month

byok_usage decimal - Total external BYOK usage (in USD) for the API key

byok_usage_daily decimal - External BYOK usage (in USD) for the current UTC day

byok_usage_weekly decimal - External BYOK usage (in USD) for the current UTC week (Monday-Sunday)

byok_usage_monthly decimal - External BYOK usage (in USD) for current UTC month

created_at string - ISO 8601 timestamp of when the API key was created

updated_at string - ISO 8601 timestamp of when the API key was last updated

expires_at? string - ISO 8601 UTC timestamp when the API key expires, or null if no expiration

openrouter: AssignKeysRequest

Fields

key_hashes AssignKeyHash[] - Array of API key hashes to assign to the guardrail

openrouter: AssignKeysResponse

Fields

assigned_count decimal - Number of keys successfully assigned

openrouter: AssignMembersRequest

Fields

member_user_ids AssignMemberUserId[] - Array of member user IDs to assign to the guardrail

openrouter: AssignMembersResponse

Fields

assigned_count decimal - Number of members successfully assigned

openrouter: AssistantMessage

Assistant message for requests and responses

Fields

role "assistant" -

content? anydata - Assistant message content

name? string - Optional name for the assistant

tool_calls? ChatMessageToolCall[] - Tool calls made by the assistant

refusal? string - Refusal message if content was refused

reasoning? string - Reasoning output

reasoning_details? AssistantMessageReasoningDetails - Reasoning details for extended thinking models

images? AssistantMessageImages - Generated images from image generation models

audio? ChatCompletionAudioOutput - Audio output data or reference

phase? anydata - The phase of an assistant message. Use commentary for an intermediate assistant message and final_answer for the final assistant message. For follow-up requests with models like gpt-5.3-codex and later, preserve and resend phase on all assistant messages. Omitting it can degrade performance. Not used for user messages.

openrouter: AuthKeyRequest

Fields

code string - The authorization code received from the OAuth redirect

code_verifier? string - The code verifier if code_challenge was used in the authorization request

code_challenge_method? "S256"|"plain" - The method used to generate the code challenge

openrouter: AuthKeyResponse

Fields

'key string - The API key to use for OpenRouter requests

user_id string - User ID associated with the API key

openrouter: BaseAnthropicMessagesResponse

Fields

id string -

'type "message" -

role "assistant" -

content (anydata)[] -

model string -

stop_reason "end_turn"|"max_tokens"|"stop_sequence"|"tool_use"|"pause_turn"|"refusal" -

stop_sequence string -

usage BaseAnthropicMessagesResponse_usage -

openrouter: BaseAnthropicMessagesResponse_usage

Fields

input_tokens decimal -

output_tokens decimal -

cache_creation_input_tokens decimal -

cache_read_input_tokens decimal -

cache_creation BaseAnthropicMessagesResponse_usage_cache_creation -

inference_geo string -

server_tool_use BaseAnthropicMessagesResponse_usage_server_tool_use -

service_tier "standard"|"priority"|"batch" -

openrouter: BaseAnthropicMessagesResponse_usage_cache_creation

Fields

ephemeral_5m_input_tokens decimal -

ephemeral_1h_input_tokens decimal -

openrouter: BaseAnthropicMessagesResponse_usage_server_tool_use

Fields

web_search_requests decimal -

openrouter: BulkAssignKeysToGuardrailHeaders

Represents the Headers record for the operation: bulkAssignKeysToGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: BulkAssignMembersToGuardrailHeaders

Represents the Headers record for the operation: bulkAssignMembersToGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: BulkUnassignKeysFromGuardrailHeaders

Represents the Headers record for the operation: bulkUnassignKeysFromGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: BulkUnassignMembersFromGuardrailHeaders

Represents the Headers record for the operation: bulkUnassignMembersFromGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ChatCompletionAudioOutput

Audio output data or reference

Fields

id? string - Audio output identifier

expires_at? decimal - Audio expiration timestamp

data? string - Base64 encoded audio data

transcript? string - Audio transcript

openrouter: ChatGenerationParams

Chat completion request parameters

Fields

provider? AnthropicMessagesProvider - When multiple model providers are available, optionally indicate your routing preference.

plugins? (record { id "auto-router" , enabled boolean, allowed_models string[] }|record { id "moderation" }|record { id "web" , enabled boolean, max_results decimal, search_prompt string, engine WebSearchEngine, native_tool_version "web_search_20250305"|"web_search_20260209" , include_domains string[], exclude_domains string[] }|record { id "file-parser" , enabled boolean, pdf PDFParserOptions }|record { id "response-healing" , enabled boolean })[] - Plugins you want to enable for this request, including their settings.

route? "fallback"|"sort" - DEPRECATED Use providers.sort.partition instead. Backwards-compatible alias for providers.sort.partition. Accepts legacy values: "fallback" (maps to "model"), "sort" (maps to "none").

user? string - Unique user identifier

session_id? string - A unique identifier for grouping related requests (e.g., a conversation or agent workflow) for observability. If provided in both the request body and the x-session-id header, the body value takes precedence. Maximum of 128 characters.

trace? OpenResponsesRequest_trace - Metadata for observability and tracing. Known keys (trace_id, trace_name, span_name, generation_name, parent_span_id) have special handling. Additional keys are passed through as custom metadata to configured broadcast destinations.

messages Message[] - List of messages for the conversation

model? ModelName - Model to use for completion

models? ModelNames - Models to use for completion

frequency_penalty? decimal - Frequency penalty (-2.0 to 2.0)

logit_bias? record {||} - Token logit bias adjustments

logprobs? boolean - Return log probabilities

top_logprobs? decimal - Number of top log probabilities to return (0-20)

max_completion_tokens? decimal - Maximum tokens in completion

max_tokens? decimal - Maximum tokens (deprecated, use max_completion_tokens). Note: some providers enforce a minimum of 16.

metadata? record { string... } - Key-value pairs for additional object information (max 16 pairs, 64 char keys, 512 char values)

presence_penalty? decimal - Presence penalty (-2.0 to 2.0)

reasoning? ChatGenerationParams_reasoning - Configuration options for reasoning models

response_format? ResponseFormatText|ResponseFormatJSONObject|ResponseFormatJSONSchema|ResponseFormatTextGrammar|ResponseFormatTextPython - Response format configuration

seed? int - Random seed for deterministic outputs

stop? anydata - Stop sequences (up to 4)

'stream boolean(default false) - Enable streaming response

stream_options? ChatStreamOptions - Streaming configuration options

temperature decimal(default 1) - Sampling temperature (0-2)

parallel_tool_calls? boolean -

tool_choice? ToolChoiceOption - Tool choice configuration

tools? ToolDefinitionJson[] - Available tools for function calling

top_p decimal(default 1) - Nucleus sampling parameter (0-1)

debug? DebugOptions - Debug options for inspecting request transformations (streaming only)

image_config? record {} - Provider-specific image configuration options. Keys and values vary by model/provider. See https://openrouter.ai/docs/guides/overview/multimodal/image-generation for more details.

modalities? ("text"|"image"|"audio")[] - Output modalities for the response. Supported values are "text", "image", and "audio".

cache_control? ChatGenerationParams_cache_control - Enable automatic prompt caching. When set, the system automatically applies cache breakpoints to the last cacheable block in the request. Currently supported for Anthropic Claude models.

openrouter: ChatGenerationParams_cache_control

Enable automatic prompt caching. When set, the system automatically applies cache breakpoints to the last cacheable block in the request. Currently supported for Anthropic Claude models.

Fields

'type "ephemeral" -

ttl? "5m"|"1h" -

openrouter: ChatGenerationParams_reasoning

Configuration options for reasoning models

Fields

effort? "xhigh"|"high"|"medium"|"low"|"minimal"|"none" - Constrains effort on reasoning for reasoning models

summary? anydata -

openrouter: ChatGenerationTokenUsage

Token usage statistics

Fields

completion_tokens decimal - Number of tokens in the completion

prompt_tokens decimal - Number of tokens in the prompt

total_tokens decimal - Total number of tokens

completion_tokens_details? ChatGenerationTokenUsage_completion_tokens_details - Detailed completion token usage

prompt_tokens_details? ChatGenerationTokenUsage_prompt_tokens_details - Detailed prompt token usage

openrouter: ChatGenerationTokenUsage_completion_tokens_details

Detailed completion token usage

Fields

reasoning_tokens? decimal - Tokens used for reasoning

audio_tokens? decimal - Tokens used for audio output

accepted_prediction_tokens? decimal - Accepted prediction tokens

rejected_prediction_tokens? decimal - Rejected prediction tokens

openrouter: ChatGenerationTokenUsage_prompt_tokens_details

Detailed prompt token usage

Fields

cached_tokens? decimal - Cached prompt tokens

cache_write_tokens? decimal - Tokens written to cache. Only returned for models with explicit caching and cache write pricing.

audio_tokens? decimal - Audio input tokens

video_tokens? decimal - Video input tokens

openrouter: ChatMessageContentItemAudio

Audio input content part. Supported audio formats vary by provider.

Fields

'type "input_audio" -

input_audio ChatMessageContentItemAudio_input_audio -

openrouter: ChatMessageContentItemAudio_input_audio

Fields

data string - Base64 encoded audio data

format string - Audio format (e.g., wav, mp3, flac, m4a, ogg, aiff, aac, pcm16, pcm24). Supported formats vary by provider.

openrouter: ChatMessageContentItemCacheControl

Cache control for the content part

Fields

'type "ephemeral" -

ttl? "5m"|"1h" -

openrouter: ChatMessageContentItemFile

File content part for document processing

Fields

'type "file" -

file ChatMessageContentItemFile_file -

openrouter: ChatMessageContentItemFile_file

Fields

file_data? string - File content as base64 data URL or URL

file_id? string - File ID for previously uploaded files

filename? string - Original filename

openrouter: ChatMessageContentItemImage

Image content part for vision models

Fields

'type "image_url" -

image_url ChatMessageContentItemImage_image_url -

openrouter: ChatMessageContentItemImage_image_url

Fields

url string - URL of the image (data: URLs supported)

detail? "auto"|"low"|"high" - Image detail level for vision models

openrouter: ChatMessageContentItemText

Text content part

Fields

'type "text" -

text string -

cache_control? ChatMessageContentItemCacheControl - Cache control for the content part

openrouter: ChatMessageContentItemVideo

Video input content part

Fields

'type "video_url" -

video_url VideoInput - Video input object

openrouter: ChatMessageContentItemVideoLegacy

Deprecated

Video input content part (legacy format - deprecated)

Fields

'type "input_video" -

video_url VideoInput - Video input object

Deprecated

openrouter: ChatMessageTokenLogprob

Token log probability information

Fields

token string - The token

logprob decimal - Log probability of the token

bytes decimal[] - UTF-8 bytes of the token

top_logprobs record { token string, logprob decimal, bytes decimal[] }[] - Top alternative tokens with probabilities

openrouter: ChatMessageTokenLogprobs

Log probabilities for the completion

Fields

content ChatMessageTokenLogprob[] - Log probabilities for content tokens

refusal? ChatMessageTokenLogprob[] - Log probabilities for refusal tokens

openrouter: ChatMessageToolCall

Tool call made by the assistant

Fields

id string - Tool call identifier

'type "function" -

'function ChatMessageToolCall_function -

openrouter: ChatMessageToolCall_function

Fields

name string - Function name to call

arguments string - Function arguments as JSON string

openrouter: ChatResponse

Chat completion response

Fields

id string - Unique completion identifier

choices ChatResponseChoice[] - List of completion choices

created decimal - Unix timestamp of creation

model string - Model used for completion

'object "chat.completion" -

system_fingerprint? string - System fingerprint

usage? ChatGenerationTokenUsage - Token usage statistics

openrouter: ChatResponseChoice

Chat completion choice

Fields

finish_reason anydata -

index decimal - Choice index

message AssistantMessage - Assistant message for requests and responses

logprobs? ChatMessageTokenLogprobs - Log probabilities for the completion

openrouter: ChatStreamOptions

Streaming configuration options

Fields

include_usage? boolean - Deprecated: This field has no effect. Full usage details are always included.

openrouter: CoinbaseCreditData

Fields

id string -

created_at string -

expires_at string -

web3_data Web3Data -

openrouter: CoinbaseCreditResponse

Fields

data CoinbaseCreditData -

openrouter: CompoundFilter

A compound filter that combines multiple comparison or compound filters

Fields

'type "and"|"or" -

filters record {}[] -

openrouter: ConnectionConfig

Closed record

Provides a set of configurations for controlling the behaviours when communicating with a remote HTTP endpoint.

Fields

auth BearerTokenConfig - Configurations related to client authentication

httpVersion HttpVersion(default http:HTTP_2_0) - The HTTP version understood by the client

http1Settings ClientHttp1Settings(default {}) - Configurations related to HTTP/1.x protocol

http2Settings ClientHttp2Settings(default {}) - Configurations related to HTTP/2 protocol

timeout decimal(default 30) - The maximum time to wait (in seconds) for a response before closing the connection

forwarded string(default "disable") - The choice of setting forwarded/x-forwarded header

followRedirects? FollowRedirects - Configurations associated with Redirection

poolConfig? PoolConfiguration - Configurations associated with request pooling

cache CacheConfig(default {}) - HTTP caching related configurations

compression Compression(default http:COMPRESSION_AUTO) - Specifies the way of handling compression (accept-encoding) header

circuitBreaker? CircuitBreakerConfig - Configurations associated with the behaviour of the Circuit Breaker

retryConfig? RetryConfig - Configurations associated with retrying

cookieConfig? CookieConfig - Configurations associated with cookies

responseLimits ResponseLimitConfigs(default {}) - Configurations associated with inbound response size limits

secureSocket? ClientSecureSocket - SSL/TLS-related options

proxy? ProxyConfig - Proxy server related options

socketConfig ClientSocketConfig(default {}) - Provides settings related to client socket configuration

validation boolean(default true) - Enables the inbound payload validation functionality which provided by the constraint package. Enabled by default

laxDataBinding boolean(default true) - Enables relaxed data binding on the client side. When enabled, nil values are treated as optional, and absent fields are handled as nilable types. Enabled by default.

openrouter: CreateAuthKeysCodeHeaders

Represents the Headers record for the operation: createAuthKeysCode

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreateChargeRequest

Create a Coinbase charge for crypto payment

Fields

amount decimal -

sender string -

chain_id 1|137|8453 -

openrouter: CreateCoinbaseChargeHeaders

Represents the Headers record for the operation: createCoinbaseCharge

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreatedKeyData

The created API key information

Fields

hash string - Unique hash identifier for the API key

name string - Name of the API key

label string - Human-readable label for the API key

disabled boolean - Whether the API key is disabled

'limit decimal - Spending limit for the API key in USD

limit_remaining decimal - Remaining spending limit in USD

limit_reset string - Type of limit reset for the API key

include_byok_in_limit boolean - Whether to include external BYOK usage in the credit limit

usage decimal - Total OpenRouter credit usage (in USD) for the API key

usage_daily decimal - OpenRouter credit usage (in USD) for the current UTC day

usage_weekly decimal - OpenRouter credit usage (in USD) for the current UTC week (Monday-Sunday)

usage_monthly decimal - OpenRouter credit usage (in USD) for the current UTC month

byok_usage decimal - Total external BYOK usage (in USD) for the API key

byok_usage_daily decimal - External BYOK usage (in USD) for the current UTC day

byok_usage_weekly decimal - External BYOK usage (in USD) for the current UTC week (Monday-Sunday)

byok_usage_monthly decimal - External BYOK usage (in USD) for current UTC month

created_at string - ISO 8601 timestamp of when the API key was created

updated_at string - ISO 8601 timestamp of when the API key was last updated

expires_at? string - ISO 8601 UTC timestamp when the API key expires, or null if no expiration

openrouter: CreateEmbeddingsHeaders

Represents the Headers record for the operation: createEmbeddings

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreateEmbeddingsRequest

Fields

input string|string[]|decimal[]|decimal[][]|record { content (record { 'type "text" , text string }|record { 'type "image_url" , image_url record { url string } })[] }[] -

model string -

encoding_format? "float"|"base64" -

dimensions? int -

user? string -

provider? ProviderPreferences - Provider routing preferences for the request.

input_type? string -

openrouter: CreateEmbeddingsResponse

Fields

id? string -

'object "list" -

data record { 'object "embedding" , embedding decimal[]|string, index decimal }[] -

model string -

usage? CreateEmbeddingsResponseUsage -

openrouter: CreateEmbeddingsResponseUsage

Fields

prompt_tokens decimal -

total_tokens decimal -

cost? decimal -

openrouter: CreateGuardrailHeaders

Represents the Headers record for the operation: createGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreateGuardrailRequest

Fields

name string - Name for the new guardrail

description? string - Description of the guardrail

limit_usd? decimal - Spending limit in USD

reset_interval? "daily"|"weekly"|"monthly" - Interval at which the limit resets (daily, weekly, monthly)

allowed_providers? string[] - List of allowed provider IDs

ignored_providers? string[] - List of provider IDs to exclude from routing

allowed_models? string[] - Array of model identifiers (slug or canonical_slug accepted)

enforce_zdr? boolean - Whether to enforce zero data retention

openrouter: CreateGuardrailResponse

Fields

data GuardrailData - The created guardrail

openrouter: CreateKeyRequest

Fields

name string - Name for the new API key

'limit? decimal - Optional spending limit for the API key in USD

limit_reset? "daily"|"weekly"|"monthly" - Type of limit reset for the API key (daily, weekly, monthly, or null for no reset). Resets happen automatically at midnight UTC, and weeks are Monday through Sunday.

include_byok_in_limit? boolean - Whether to include BYOK usage in the limit

expires_at? string - Optional ISO 8601 UTC timestamp when the API key should expire. Must be UTC, other timezones will be rejected

openrouter: CreateKeyResponse

Fields

data CreatedKeyData - The created API key information

'key string - The actual API key string (only shown once)

openrouter: CreateKeysHeaders

Represents the Headers record for the operation: createKeys

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreateMessagesHeaders

Represents the Headers record for the operation: createMessages

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreateResponsesHeaders

Represents the Headers record for the operation: createResponses

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: CreditsData

Fields

total_credits decimal - Total credits purchased

total_usage decimal - Total credits used

openrouter: CreditsResponse

Total credits purchased and used

Fields

data CreditsData -

openrouter: CurrentKeyData

Current API key information

Fields

label string - Human-readable label for the API key

'limit decimal - Spending limit for the API key in USD

usage decimal - Total OpenRouter credit usage (in USD) for the API key

usage_daily decimal - OpenRouter credit usage (in USD) for the current UTC day

usage_weekly decimal - OpenRouter credit usage (in USD) for the current UTC week (Monday-Sunday)

usage_monthly decimal - OpenRouter credit usage (in USD) for the current UTC month

byok_usage decimal - Total external BYOK usage (in USD) for the API key

byok_usage_daily decimal - External BYOK usage (in USD) for the current UTC day

byok_usage_weekly decimal - External BYOK usage (in USD) for the current UTC week (Monday-Sunday)

byok_usage_monthly decimal - External BYOK usage (in USD) for current UTC month

is_free_tier boolean - Whether this is a free tier API key

is_management_key boolean - Whether this is a management key

is_provisioning_key boolean - Whether this is a management key

limit_remaining decimal - Remaining spending limit in USD

limit_reset string - Type of limit reset for the API key

include_byok_in_limit boolean - Whether to include external BYOK usage in the credit limit

expires_at? string - ISO 8601 UTC timestamp when the API key expires, or null if no expiration

rate_limit KeyRateLimit - Legacy rate limit information about a key. Will always return -1.

openrouter: CurrentKeyResponse

Fields

data CurrentKeyData - Current API key information

openrouter: DebugOptions

Debug options for inspecting request transformations (streaming only)

Fields

echo_upstream_body? boolean - If true, includes the transformed upstream request body in a debug chunk at the start of the stream. Only works with streaming mode.

openrouter: DefaultParameters

Closed record

Default parameters for this model

Fields

temperature? decimal -

top_p? decimal -

top_k? int -

frequency_penalty? decimal -

presence_penalty? decimal -

repetition_penalty? decimal -

openrouter: DeleteGuardrailHeaders

Represents the Headers record for the operation: deleteGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: DeleteGuardrailResponse

Fields

deleted boolean - Confirmation that the guardrail was deleted

openrouter: DeleteKeyResponse

Fields

deleted boolean - Confirmation that the API key was deleted

openrouter: DeleteKeysHeaders

Represents the Headers record for the operation: deleteKeys

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: DeveloperMessage

Developer message

Fields

role "developer" -

content string|ChatMessageContentItemText[] - Developer message content

name? string - Optional name for the developer message

openrouter: ExchangeAuthCodeForAPIKeyHeaders

Represents the Headers record for the operation: exchangeAuthCodeForAPIKey

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: FileCitation

Fields

'type "file_citation" -

file_id string -

filename string -

index decimal -

openrouter: FilePath

Fields

'type "file_path" -

file_id string -

index decimal -

openrouter: GenerationData

Generation data

Fields

id string - Unique identifier for the generation

upstream_id string - Upstream provider's identifier for this generation

total_cost decimal - Total cost of the generation in USD

cache_discount decimal - Discount applied due to caching

upstream_inference_cost decimal - Cost charged by the upstream provider

created_at string - ISO 8601 timestamp of when the generation was created

model string - Model used for the generation

app_id decimal - ID of the app that made the request

streamed boolean - Whether the response was streamed

cancelled boolean - Whether the generation was cancelled

provider_name string - Name of the provider that served the request

latency decimal - Total latency in milliseconds

moderation_latency decimal - Moderation latency in milliseconds

generation_time decimal - Time taken for generation in milliseconds

finish_reason string - Reason the generation finished

tokens_prompt decimal - Number of tokens in the prompt

tokens_completion decimal - Number of tokens in the completion

native_tokens_prompt decimal - Native prompt tokens as reported by provider

native_tokens_completion decimal - Native completion tokens as reported by provider

native_tokens_completion_images decimal - Native completion image tokens as reported by provider

native_tokens_reasoning decimal - Native reasoning tokens as reported by provider

native_tokens_cached decimal - Native cached tokens as reported by provider

num_media_prompt decimal - Number of media items in the prompt

num_input_audio_prompt decimal - Number of audio inputs in the prompt

num_media_completion decimal - Number of media items in the completion

num_search_results decimal - Number of search results included

origin string - Origin URL of the request

usage decimal - Usage amount in USD

is_byok boolean - Whether this used bring-your-own-key

native_finish_reason string - Native finish reason as reported by provider

external_user string - External user identifier

api_type "completions"|"embeddings" - Type of API used for the generation

router string - Router used for the request (e.g., openrouter/auto)

provider_responses record { id string, endpoint_id string, model_permaslug string, provider_name "AnyScale"|"Atoma"|"Cent-ML"|"CrofAI"|"Enfer"|"GoPomelo"|"HuggingFace"|"Hyperbolic 2"|"InoCloud"|"Kluster"|"Lambda"|"Lepton"|"Lynn 2"|"Lynn"|"Mancer"|"Meta"|"Modal"|"Nineteen"|"OctoAI"|"Recursal"|"Reflection"|"Replicate"|"SambaNova 2"|"SF Compute"|"Targon"|"Together 2"|"Ubicloud"|"01.AI"|"AI21"|"AionLabs"|"Alibaba"|"Ambient"|"Amazon Bedrock"|"Amazon Nova"|"Anthropic"|"Arcee AI"|"AtlasCloud"|"Avian"|"Azure"|"BaseTen"|"BytePlus"|"Black Forest Labs"|"Cerebras"|"Chutes"|"Cirrascale"|"Clarifai"|"Cloudflare"|"Cohere"|"Crusoe"|"DeepInfra"|"DeepSeek"|"Featherless"|"Fireworks"|"Friendli"|"GMICloud"|"Google"|"Google AI Studio"|"Groq"|"Hyperbolic"|"Inception"|"Inceptron"|"InferenceNet"|"Ionstream"|"Infermatic"|"Io Net"|"Inflection"|"Liquid"|"Mara"|"Mancer 2"|"Minimax"|"ModelRun"|"Mistral"|"Modular"|"Moonshot AI"|"Morph"|"NCompass"|"Nebius"|"NextBit"|"Novita"|"Nvidia"|"OpenAI"|"OpenInference"|"Parasail"|"Perplexity"|"Phala"|"Relace"|"SambaNova"|"Seed"|"SiliconFlow"|"Sourceful"|"StepFun"|"Stealth"|"StreamLake"|"Switchpoint"|"Together"|"Upstage"|"Venice"|"WandB"|"Xiaomi"|"xAI"|"Z.AI"|"FakeProvider" , status decimal, latency decimal, is_byok boolean }[] - List of provider responses for this generation, including fallback attempts

user_agent string - User-Agent header from the request

http_referer string - Referer header from the request

openrouter: GenerationResponse

Generation response

Fields

data GenerationData - Generation data

openrouter: GetCreditsHeaders

Represents the Headers record for the operation: getCredits

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetCurrentKeyHeaders

Represents the Headers record for the operation: getCurrentKey

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetGenerationHeaders

Represents the Headers record for the operation: getGeneration

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetGenerationQueries

Represents the Queries record for the operation: getGeneration

Fields

id string -

openrouter: GetGuardrailHeaders

Represents the Headers record for the operation: getGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetGuardrailResponse

Fields

data GuardrailInfo - The guardrail

openrouter: GetKeyHeaders

Represents the Headers record for the operation: getKey

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetModelsHeaders

Represents the Headers record for the operation: getModels

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetModelsQueries

Represents the Queries record for the operation: getModels

Fields

category? "programming"|"roleplay"|"marketing"|"marketing/seo"|"technology"|"science"|"translation"|"legal"|"finance"|"health"|"trivia"|"academia" - Filter models by use case category

supported_parameters? string -

openrouter: GetUserActivityHeaders

Represents the Headers record for the operation: getUserActivity

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: GetUserActivityQueries

Represents the Queries record for the operation: getUserActivity

Fields

date? string - Filter by a single UTC date in the last 30 days (YYYY-MM-DD format).

openrouter: GuardrailData

The created guardrail

Fields

id string - Unique identifier for the guardrail

name string - Name of the guardrail

description? string - Description of the guardrail

limit_usd? decimal - Spending limit in USD

reset_interval? "daily"|"weekly"|"monthly" - Interval at which the limit resets (daily, weekly, monthly)

allowed_providers? string[] - List of allowed provider IDs

ignored_providers? string[] - List of provider IDs to exclude from routing

allowed_models? string[] - Array of model canonical_slugs (immutable identifiers)

enforce_zdr? boolean - Whether to enforce zero data retention

created_at string - ISO 8601 timestamp of when the guardrail was created

updated_at? string - ISO 8601 timestamp of when the guardrail was last updated

openrouter: GuardrailInfo

The guardrail

Fields

id string - Unique identifier for the guardrail

name string - Name of the guardrail

description? string - Description of the guardrail

limit_usd? decimal - Spending limit in USD

reset_interval? "daily"|"weekly"|"monthly" - Interval at which the limit resets (daily, weekly, monthly)

allowed_providers? string[] - List of allowed provider IDs

ignored_providers? string[] - List of provider IDs to exclude from routing

allowed_models? string[] - Array of model canonical_slugs (immutable identifiers)

enforce_zdr? boolean - Whether to enforce zero data retention

created_at string - ISO 8601 timestamp of when the guardrail was created

updated_at? string - ISO 8601 timestamp of when the guardrail was last updated

openrouter: GuardrailKeyAssignmentsResponse

Fields

data record { id string, key_hash string, guardrail_id string, key_name string, key_label string, assigned_by string, created_at string }[] - List of key assignments

total_count decimal - Total number of key assignments for this guardrail

openrouter: GuardrailListItem

Fields

id string -

name string -

description? string -

limit_usd? decimal -

reset_interval? "daily"|"weekly"|"monthly" -

allowed_providers? string[] -

ignored_providers? string[] -

allowed_models? string[] -

enforce_zdr? boolean -

created_at string -

updated_at? string -

openrouter: GuardrailListResponse

Fields

data record { id string, name string, description string, limit_usd decimal, reset_interval "daily"|"weekly"|"monthly" , allowed_providers string[], ignored_providers string[], allowed_models string[], enforce_zdr boolean, created_at string, updated_at string }[] - List of guardrails

total_count decimal - Total number of guardrails

openrouter: GuardrailMemberAssignmentsResponse

Fields

data record { id string, user_id string, organization_id string, guardrail_id string, assigned_by string, created_at string }[] - List of member assignments

total_count decimal - Total number of member assignments

openrouter: JSONSchemaConfig

JSON Schema configuration object

Fields

name string - Schema name (a-z, A-Z, 0-9, underscores, dashes, max 64 chars)

description? string - Schema description for the model

schema? record {} - JSON Schema object

strict? boolean - Enable strict schema adherence

openrouter: KeyCodeData

Auth code data

Fields

id string - The authorization code ID to use in the exchange request

app_id decimal - The application ID associated with this auth code

created_at string - ISO 8601 timestamp of when the auth code was created

openrouter: KeyCodeRequest

Fields

callback_url string - The callback URL to redirect to after authorization. Note, only https URLs on ports 443 and 3000 are allowed.

code_challenge? string - PKCE code challenge for enhanced security

code_challenge_method? "S256"|"plain" - The method used to generate the code challenge

'limit? decimal - Credit limit for the API key to be created

expires_at? string - Optional expiration time for the API key to be created

key_label? string - Optional custom label for the API key. Defaults to the app name if not provided.

usage_limit_type? "daily"|"weekly"|"monthly" - Optional credit limit reset interval. When set, the credit limit resets on this interval.

spawn_agent? string - Agent identifier for spawn telemetry

spawn_cloud? string - Cloud identifier for spawn telemetry

openrouter: KeyCodeResponse

Fields

data KeyCodeData - Auth code data

openrouter: KeyDetailsResponse

Fields

data ApiKeyInfo - The API key information

openrouter: KeyRateLimit

Deprecated

Legacy rate limit information about a key. Will always return -1.

Fields

requests decimal - Number of requests allowed per interval

interval string - Rate limit interval

note string - Note about the rate limit

Deprecated

openrouter: KeysListResponse

Fields

data record { hash string, name string, label string, disabled boolean, 'limit decimal, limit_remaining decimal, limit_reset string, include_byok_in_limit boolean, usage decimal, usage_daily decimal, usage_weekly decimal, usage_monthly decimal, byok_usage decimal, byok_usage_daily decimal, byok_usage_weekly decimal, byok_usage_monthly decimal, created_at string, updated_at string, expires_at string }[] - List of API keys

openrouter: ListEmbeddingsModelsHeaders

Represents the Headers record for the operation: listEmbeddingsModels

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListEndpointsHeaders

Represents the Headers record for the operation: listEndpoints

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListEndpointsResponse

List of available endpoints for a model

Fields

id string - Unique identifier for the model

name string - Display name of the model

created decimal - Unix timestamp of when the model was created

description string - Description of the model

architecture record { tokenizer ModelGroup, instruct_type InstructType, modality string, input_modalities InputModality[], output_modalities OutputModality[] } -

endpoints PublicEndpoint[] - List of available endpoints for this model

openrouter: ListEndpointsZdrHeaders

Represents the Headers record for the operation: listEndpointsZdr

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListGuardrailKeyAssignmentsHeaders

Represents the Headers record for the operation: listGuardrailKeyAssignments

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListGuardrailKeyAssignmentsQueries

Represents the Queries record for the operation: listGuardrailKeyAssignments

Fields

offset? string - Number of records to skip for pagination

'limit? string - Maximum number of records to return (max 100)

openrouter: ListGuardrailMemberAssignmentsHeaders

Represents the Headers record for the operation: listGuardrailMemberAssignments

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListGuardrailMemberAssignmentsQueries

Represents the Queries record for the operation: listGuardrailMemberAssignments

Fields

offset? string - Number of records to skip for pagination

'limit? string - Maximum number of records to return (max 100)

openrouter: ListGuardrailsHeaders

Represents the Headers record for the operation: listGuardrails

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListGuardrailsQueries

Represents the Queries record for the operation: listGuardrails

Fields

offset? string - Number of records to skip for pagination

'limit? string - Maximum number of records to return (max 100)

openrouter: ListHeaders

Represents the Headers record for the operation: list

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListKeyAssignmentsHeaders

Represents the Headers record for the operation: listKeyAssignments

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListKeyAssignmentsQueries

Represents the Queries record for the operation: listKeyAssignments

Fields

offset? string - Number of records to skip for pagination

'limit? string - Maximum number of records to return (max 100)

openrouter: ListMemberAssignmentsHeaders

Represents the Headers record for the operation: listMemberAssignments

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListMemberAssignmentsQueries

Represents the Queries record for the operation: listMemberAssignments

Fields

offset? string - Number of records to skip for pagination

'limit? string - Maximum number of records to return (max 100)

openrouter: ListModelsCountHeaders

Represents the Headers record for the operation: listModelsCount

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListModelsUserHeaders

Represents the Headers record for the operation: listModelsUser

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListProvidersHeaders

Represents the Headers record for the operation: listProviders

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: ListQueries

Represents the Queries record for the operation: list

Fields

include_disabled? string - Whether to include disabled API keys in the response

offset? string - Number of API keys to skip for pagination

openrouter: Model

Information about an AI model available on OpenRouter

Fields

id string - Unique identifier for the model

canonical_slug string - Canonical slug for the model

hugging_face_id? string - Hugging Face model identifier, if applicable

name string - Display name of the model

created decimal - Unix timestamp of when the model was created

description? string - Description of the model

pricing PublicPricing - Pricing information for the model

context_length decimal - Maximum context length in tokens

architecture ModelArchitecture - Model architecture information

top_provider TopProviderInfo - Information about the top provider for this model

per_request_limits PerRequestLimits - Per-request token limits

supported_parameters Parameter[] - List of supported parameters for this model

default_parameters DefaultParameters - Default parameters for this model

expiration_date? string - The date after which the model may be removed. ISO 8601 date string (YYYY-MM-DD) or null if no expiration.

openrouter: ModelArchitecture

Model architecture information

Fields

tokenizer? ModelGroup - Tokenizer type used by the model

instruct_type? "none"|"airoboros"|"alpaca"|"alpaca-modif"|"chatml"|"claude"|"code-llama"|"gemma"|"llama2"|"llama3"|"mistral"|"nemotron"|"neural"|"openchat"|"phi3"|"rwkv"|"vicuna"|"zephyr"|"deepseek-r1"|"deepseek-v3.1"|"qwq"|"qwen3" - Instruction format type

modality string - Primary modality of the model

input_modalities InputModality[] - Supported input modalities

output_modalities OutputModality[] - Supported output modalities

openrouter: ModelEndpointsResponse

Fields

data ListEndpointsResponse - List of available endpoints for a model

openrouter: ModelsCountResponse

Model count data

Fields

data ModelsCountResponse_data - Model count data

openrouter: ModelsCountResponse_data

Model count data

Fields

count decimal - Total number of available models

openrouter: ModelsListResponse

List of available models

Fields

data ModelsListResponseData - List of available models

openrouter: NamedToolChoice

Named tool choice for specific function

Fields

'type "function" -

'function NamedToolChoice_function -

openrouter: NamedToolChoice_function

Fields

name string - Function name to call

openrouter: OpenAIResponsesIncompleteDetails

Fields

reason? "max_output_tokens"|"content_filter" -

openrouter: OpenAIResponsesNonStreamingResponse

Fields

id string -

'object "response" -

created_at decimal -

model string -

status OpenAIResponsesResponseStatus -

completed_at decimal -

output (OutputMessage|OutputItemReasoning|OutputItemFunctionCall|OutputItemWebSearchCall|OutputItemFileSearchCall|OutputItemImageGenerationCall)[] -

user? string -

output_text? string -

prompt_cache_key? string -

safety_identifier? string -

'error ResponsesErrorField - Error information returned from the API

incomplete_details OpenAIResponsesIncompleteDetails -

usage? OpenAIResponsesUsage -

max_tool_calls? decimal -

top_logprobs? decimal -

max_output_tokens? decimal -

temperature decimal -

top_p decimal -

presence_penalty decimal -

frequency_penalty decimal -

instructions OpenAIResponsesInput -

metadata OpenResponsesRequestMetadata - Metadata key-value pairs for the request. Keys must be ≤64 characters and cannot contain brackets. Values must be ≤512 characters. Maximum 16 pairs allowed.

tools (record {}|OpenResponsesWebSearchPreviewTool|OpenResponsesWebSearchPreview20250311Tool|OpenResponsesWebSearchTool|OpenResponsesWebSearch20250826Tool|OpenResponsesFileSearchTool|OpenResponsesComputerTool|OpenResponsesCodeInterpreterTool|OpenResponsesMcpTool|OpenResponsesImageGenerationTool|OpenResponsesLocalShellTool|OpenResponsesFunctionShellTool|OpenResponsesApplyPatchTool|OpenResponsesCustomTool)[] -

tool_choice OpenAIResponsesToolChoice -

parallel_tool_calls boolean -

prompt? OpenAIResponsesPrompt -

background? boolean -

previous_response_id? string -

reasoning? OpenAIResponsesReasoningConfig -

service_tier? OpenAIResponsesServiceTier -

store? boolean -

truncation? OpenAIResponsesTruncation -

text? ResponseTextConfig - Text output configuration including format and verbosity

openrouter: OpenAIResponsesPrompt

Fields

id string -

variables? record {} -

openrouter: OpenAIResponsesReasoningConfig

Fields

effort? OpenAIResponsesReasoningEffort -

summary? ReasoningSummaryVerbosity -

openrouter: OpenAIResponsesRefusalContent

Fields

'type "refusal" -

refusal string -

openrouter: OpenAIResponsesUsage

Fields

input_tokens decimal -

input_tokens_details OpenAIResponsesUsage_input_tokens_details -

output_tokens decimal -

output_tokens_details OpenAIResponsesUsage_output_tokens_details -

total_tokens decimal -

openrouter: OpenAIResponsesUsage_input_tokens_details

Fields

cached_tokens decimal -

openrouter: OpenAIResponsesUsage_output_tokens_details

Fields

reasoning_tokens decimal -

openrouter: OpenResponsesApplyPatchTool

Apply patch tool configuration

Fields

'type "apply_patch" -

openrouter: OpenResponsesCodeInterpreterTool

Code interpreter tool configuration

Fields

'type "code_interpreter" -

container string|record { 'type "auto" , file_ids string[], memory_limit "1g"|"4g"|"16g"|"64g" } -

openrouter: OpenResponsesComputerTool

Computer use preview tool configuration

Fields

'type "computer_use_preview" -

display_height decimal -

display_width decimal -

environment "windows"|"mac"|"linux"|"ubuntu"|"browser" -

openrouter: OpenResponsesCustomTool

Custom tool configuration

Fields

'type "custom" -

name string -

description? string -

format? record { 'type "text" }|record { 'type "grammar" , definition string, syntax "lark"|"regex" } -

openrouter: OpenResponsesEasyInputMessage

Fields

'type? "message" -

role "user"|"system"|"assistant"|"developer" -

content? anydata -

phase? anydata - The phase of an assistant message. Use commentary for an intermediate assistant message and final_answer for the final assistant message. For follow-up requests with models like gpt-5.3-codex and later, preserve and resend phase on all assistant messages. Omitting it can degrade performance. Not used for user messages.

openrouter: OpenResponsesFileSearchTool

File search tool configuration

Fields

'type "file_search" -

vector_store_ids string[] -

filters? anydata -

max_num_results? int -

ranking_options? OpenResponsesFileSearchTool_ranking_options -

openrouter: OpenResponsesFileSearchTool_ranking_options

Fields

ranker? "auto"|"default-2024-11-15" -

score_threshold? decimal -

openrouter: OpenResponsesFunctionCallOutput

The output from a function call execution

Fields

'type "function_call_output" -

id? string -

call_id string -

output string|(ResponseInputText|record {}|ResponseInputFile)[] -

status? ToolCallStatus -

openrouter: OpenResponsesFunctionShellTool

Shell tool configuration

Fields

'type "shell" -

openrouter: OpenResponsesFunctionTool

Function tool definition

Fields

'type "function" -

name string -

description? string -

strict? boolean -

parameters record {} -

openrouter: OpenResponsesFunctionToolCall

A function call initiated by the model

Fields

'type "function_call" -

call_id string -

name string -

arguments string -

id string -

status? ToolCallStatus -

openrouter: OpenResponsesImageGenerationTool

Image generation tool configuration

Fields

'type "image_generation" -

background? "transparent"|"opaque"|"auto" -

input_fidelity? "high"|"low" -

input_image_mask? OpenResponsesImageGenerationTool_input_image_mask -

model? "gpt-image-1"|"gpt-image-1-mini" -

moderation? "auto"|"low" -

output_compression? decimal -

output_format? "png"|"webp"|"jpeg" -

partial_images? decimal -

quality? "low"|"medium"|"high"|"auto" -

size? "1024x1024"|"1024x1536"|"1536x1024"|"auto" -

openrouter: OpenResponsesImageGenerationTool_input_image_mask

Fields

image_url? string -

file_id? string -

openrouter: OpenResponsesInputMessageItem

Fields

id? string -

'type? "message" -

role "user"|"system"|"developer" -

content? (ResponseInputText|record {}|ResponseInputFile|ResponseInputAudio|ResponseInputVideo)[] -

openrouter: OpenResponsesLocalShellTool

Local shell tool configuration

Fields

'type "local_shell" -

openrouter: OpenResponsesMcpTool

MCP (Model Context Protocol) tool configuration

Fields

'type "mcp" -

server_label string -

allowed_tools? anydata -

authorization? string -

connector_id? "connector_dropbox"|"connector_gmail"|"connector_googlecalendar"|"connector_googledrive"|"connector_microsoftteams"|"connector_outlookcalendar"|"connector_outlookemail"|"connector_sharepoint" -

headers? record { string... } -

require_approval? anydata -

server_description? string -

server_url? string -

openrouter: OpenResponsesNonStreamingResponse

Complete non-streaming response from the Responses API

Fields

Fields Included from *OpenAIResponsesNonStreamingResponse

id string
object "response"
created_at decimal
model string
status OpenAIResponsesResponseStatus
completed_at decimal
output OutputMessage|OutputItemReasoning|OutputItemFunctionCall|OutputItemWebSearchCall|OutputItemFileSearchCall|OutputItemImageGenerationCall[]
user string
output_text string
prompt_cache_key string
safety_identifier string
error ResponsesErrorField
incomplete_details OpenAIResponsesIncompleteDetails
usage OpenAIResponsesUsage
max_tool_calls decimal
top_logprobs decimal
max_output_tokens decimal
temperature decimal
top_p decimal
presence_penalty decimal
frequency_penalty decimal
instructions OpenAIResponsesInput
metadata OpenResponsesRequestMetadata
tools record { type "function", name string, description string, strict boolean, parameters record { anydata... }, anydata... }|OpenResponsesWebSearchPreviewTool|OpenResponsesWebSearchPreview20250311Tool|OpenResponsesWebSearchTool|OpenResponsesWebSearch20250826Tool|OpenResponsesFileSearchTool|OpenResponsesComputerTool|OpenResponsesCodeInterpreterTool|OpenResponsesMcpTool|OpenResponsesImageGenerationTool|OpenResponsesLocalShellTool|OpenResponsesFunctionShellTool|OpenResponsesApplyPatchTool|OpenResponsesCustomTool[]
tool_choice OpenAIResponsesToolChoice
parallel_tool_calls boolean
prompt OpenAIResponsesPrompt
background boolean
previous_response_id string
reasoning OpenAIResponsesReasoningConfig
service_tier OpenAIResponsesServiceTier
store boolean
truncation OpenAIResponsesTruncation
text ResponseTextConfig
anydata...

output? ResponsesOutputItem[] -

usage? OpenResponsesUsage -

openrouter: OpenResponsesReasoning

Reasoning output item with signature and format extensions

Fields

Fields Included from *OutputItemReasoning

type "reasoning"
id string
content ReasoningTextContent[]
summary ReasoningSummaryText[]
encrypted_content string
status "completed"|"incomplete"|"in_progress"
anydata...

content? ReasoningTextContent[] -

signature? string -

format? "unknown"|"openai-responses-v1"|"azure-openai-responses-v1"|"xai-responses-v1"|"anthropic-claude-v1"|"google-gemini-v1" -

openrouter: OpenResponsesReasoningConfig

Configuration for reasoning mode in the response

Fields

Fields Included from *OpenAIResponsesReasoningConfig

effort OpenAIResponsesReasoningEffort
summary ReasoningSummaryVerbosity
anydata...

max_tokens? decimal -

enabled? boolean -

openrouter: OpenResponsesRequest

Request schema for Responses endpoint

Fields

input? OpenResponsesInput - Input for a response request - can be a string or array of items

instructions? string -

metadata? OpenResponsesRequestMetadata - Metadata key-value pairs for the request. Keys must be ≤64 characters and cannot contain brackets. Values must be ≤512 characters. Maximum 16 pairs allowed.

tools? (record {}|OpenResponsesWebSearchPreviewTool|OpenResponsesWebSearchPreview20250311Tool|OpenResponsesWebSearchTool|OpenResponsesWebSearch20250826Tool|OpenResponsesFileSearchTool|OpenResponsesComputerTool|OpenResponsesCodeInterpreterTool|OpenResponsesMcpTool|OpenResponsesImageGenerationTool|OpenResponsesLocalShellTool|OpenResponsesFunctionShellTool|OpenResponsesApplyPatchTool|OpenResponsesCustomTool)[] -

tool_choice? OpenAIResponsesToolChoice -

parallel_tool_calls? boolean -

model? string -

models? string[] -

text? OpenResponsesResponseText - Text output configuration including format and verbosity

reasoning? OpenResponsesReasoningConfig - Configuration for reasoning mode in the response

max_output_tokens? decimal -

temperature? decimal -

top_p? decimal -

top_logprobs? int -

max_tool_calls? int -

presence_penalty? decimal -

frequency_penalty? decimal -

top_k? decimal -

image_config? record {} - Provider-specific image configuration options. Keys and values vary by model/provider. See https://openrouter.ai/docs/features/multimodal/image-generation for more details.

modalities? ResponsesOutputModality[] - Output modalities for the response. Supported values are "text" and "image".

prompt_cache_key? string -

previous_response_id? string -

prompt? OpenAIResponsesPrompt -

include? OpenAIResponsesIncludable[] -

background? boolean -

safety_identifier? string -

store boolean(default false) -

service_tier "auto" (default "auto") -

truncation? OpenAIResponsesTruncation -

'stream boolean(default false) -

provider? OpenResponsesRequest_provider - When multiple model providers are available, optionally indicate your routing preference.

plugins? (record { id "auto-router" , enabled boolean, allowed_models string[] }|record { id "moderation" }|record { id "web" , enabled boolean, max_results decimal, search_prompt string, engine WebSearchEngine, native_tool_version "web_search_20250305"|"web_search_20260209" , include_domains string[], exclude_domains string[] }|record { id "file-parser" , enabled boolean, pdf PDFParserOptions }|record { id "response-healing" , enabled boolean })[] - Plugins you want to enable for this request, including their settings.

route? "fallback"|"sort" - DEPRECATED Use providers.sort.partition instead. Backwards-compatible alias for providers.sort.partition. Accepts legacy values: "fallback" (maps to "model"), "sort" (maps to "none").

user? string - A unique identifier representing your end-user, which helps distinguish between different users of your app. This allows your app to identify specific users in case of abuse reports, preventing your entire app from being affected by the actions of individual users. Maximum of 128 characters.

session_id? string - A unique identifier for grouping related requests (e.g., a conversation or agent workflow) for observability. If provided in both the request body and the x-session-id header, the body value takes precedence. Maximum of 128 characters.

trace? OpenResponsesRequest_trace - Metadata for observability and tracing. Known keys (trace_id, trace_name, span_name, generation_name, parent_span_id) have special handling. Additional keys are passed through as custom metadata to configured broadcast destinations.

openrouter: OpenResponsesRequest_provider

Closed record

When multiple model providers are available, optionally indicate your routing preference.

Fields

allow_fallbacks? boolean - Whether to allow backup providers to serve requests
- true: (default) when the primary provider (or your custom providers in "order") is unavailable, use the next best provider.
- false: use only the primary/custom provider, and return the upstream error if it's unavailable.

require_parameters? boolean - Whether to filter providers to only those that support the parameters you've provided. If this setting is omitted or set to false, then providers will receive only the parameters they support, and ignore the rest.

data_collection? DataCollection - Data collection setting. If no available model provider meets the requirement, your request will return an error.
- allow: (default) allow providers which store user data non-transiently and may train on it
- deny: use only providers which do not collect user data.

zdr? boolean - Whether to restrict routing to only ZDR (Zero Data Retention) endpoints. When true, only endpoints that do not retain prompts will be used.

enforce_distillable_text? boolean - Whether to restrict routing to only models that allow text distillation. When true, only models where the author has allowed distillation will be used.

'order? (ProviderName|string)[] - An ordered list of provider slugs. The router will attempt to use the first provider in the subset of this list that supports your requested model, and fall back to the next if it is unavailable. If no providers are available, the request will fail with an error message.

only? (ProviderName|string)[] - List of provider slugs to allow. If provided, this list is merged with your account-wide allowed provider settings for this request.

ignore? (ProviderName|string)[] - List of provider slugs to ignore. If provided, this list is merged with your account-wide ignored provider settings for this request.

quantizations? Quantization[] - A list of quantization levels to filter the provider by.

sort? anydata - The sorting strategy to use for this request, if "order" is not specified. When set, no load balancing is performed.

max_price? OpenResponsesRequest_provider_max_price - The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion.

preferred_min_throughput? PreferredMinThroughput - Preferred minimum throughput (in tokens per second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints below the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

preferred_max_latency? PreferredMaxLatency - Preferred maximum latency (in seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints above the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

openrouter: OpenResponsesRequest_provider_max_price

The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion.

Fields

prompt? BigNumberUnion - Price per million prompt tokens

completion? BigNumberUnion -

image? BigNumberUnion -

audio? BigNumberUnion -

request? BigNumberUnion -

openrouter: OpenResponsesRequest_trace

Metadata for observability and tracing. Known keys (trace_id, trace_name, span_name, generation_name, parent_span_id) have special handling. Additional keys are passed through as custom metadata to configured broadcast destinations.

Fields

trace_id? string -

trace_name? string -

span_name? string -

generation_name? string -

parent_span_id? string -

openrouter: OpenResponsesRequestMetadata

Metadata key-value pairs for the request. Keys must be ≤64 characters and cannot contain brackets. Values must be ≤512 characters. Maximum 16 pairs allowed.

Fields

string... - Rest field

openrouter: OpenResponsesResponseText

Text output configuration including format and verbosity

Fields

Fields Included from *ResponseTextConfig

format ResponseFormatTextConfig
verbosity "high"|"low"|"medium"
anydata...

openrouter: OpenResponsesUsage

Token usage information for the response

Fields

Fields Included from *OpenAIResponsesUsage

input_tokens decimal
input_tokens_details OpenAIResponsesUsage_input_tokens_details
output_tokens decimal
output_tokens_details OpenAIResponsesUsage_output_tokens_details
total_tokens decimal
anydata...

cost? decimal -

is_byok? boolean -

cost_details? record { upstream_inference_cost decimal, upstream_inference_input_cost decimal, upstream_inference_output_cost decimal } -

openrouter: OpenResponsesWebSearch20250826Tool

Web search tool configuration (2025-08-26 version)

Fields

'type "web_search_2025_08_26" -

filters? OpenResponsesWebSearchTool_filters -

search_context_size? ResponsesSearchContextSize - Size of the search context for web search tools

user_location? ResponsesWebSearchUserLocation - User location information for web search

openrouter: OpenResponsesWebSearchPreview20250311Tool

Web search preview tool configuration (2025-03-11 version)

Fields

'type "web_search_preview_2025_03_11" -

search_context_size? ResponsesSearchContextSize - Size of the search context for web search tools

user_location? WebSearchPreviewToolUserLocation -

openrouter: OpenResponsesWebSearchPreviewTool

Web search preview tool configuration

Fields

'type "web_search_preview" -

search_context_size? ResponsesSearchContextSize - Size of the search context for web search tools

user_location? WebSearchPreviewToolUserLocation -

openrouter: OpenResponsesWebSearchTool

Web search tool configuration

Fields

'type "web_search" -

filters? OpenResponsesWebSearchTool_filters -

search_context_size? ResponsesSearchContextSize - Size of the search context for web search tools

user_location? ResponsesWebSearchUserLocation - User location information for web search

openrouter: OpenResponsesWebSearchTool_filters

Fields

allowed_domains? string[] -

openrouter: OpenRouterAnthropicMessageParam

Anthropic message with OpenRouter extensions

Fields

role "user"|"assistant" -

content anydata -

openrouter: OutputItemFileSearchCall

Fields

'type "file_search_call" -

id string -

queries string[] -

status WebSearchStatus -

openrouter: OutputItemFunctionCall

Fields

'type "function_call" -

id? string -

name string -

arguments string -

call_id string -

status? "completed"|"incomplete"|"in_progress" -

openrouter: OutputItemImageGenerationCall

Fields

'type "image_generation_call" -

id string -

result? string -

status ImageGenerationStatus -

openrouter: OutputItemReasoning

Fields

'type "reasoning" -

id string -

content? ReasoningTextContent[] -

summary ReasoningSummaryText[] -

encrypted_content? string -

status? "completed"|"incomplete"|"in_progress" -

openrouter: OutputItemWebSearchCall

Fields

'type "web_search_call" -

id string -

action record { 'type "search" , query string, queries string[], sources record { 'type "url" , url string }[] }|record { 'type "open_page" , url string }|record { 'type "find_in_page" , pattern string, url string } -

status WebSearchStatus -

openrouter: OutputMessage

Fields

id string -

role "assistant" -

'type "message" -

status? "completed"|"incomplete"|"in_progress" -

content (ResponseOutputText|OpenAIResponsesRefusalContent)[] -

phase? anydata - The phase of an assistant message. Use commentary for an intermediate assistant message and final_answer for the final assistant message. For follow-up requests with models like gpt-5.3-codex and later, preserve and resend phase on all assistant messages. Omitting it can degrade performance. Not used for user messages.

openrouter: PDFParserOptions

Options for PDF parsing.

Fields

engine? PDFParserEngine - The engine to use for parsing PDF files.

openrouter: PercentileLatencyCutoffs

Percentile-based latency cutoffs. All specified cutoffs must be met for an endpoint to be preferred.

Fields

p50? decimal - Maximum p50 latency (seconds)

p75? decimal - Maximum p75 latency (seconds)

p90? decimal - Maximum p90 latency (seconds)

p99? decimal - Maximum p99 latency (seconds)

openrouter: PercentileStats

Latency percentiles in milliseconds over the last 30 minutes. Latency measures time to first token. Only visible when authenticated with an API key or cookie; returns null for unauthenticated requests.

Fields

p50 decimal - Median (50th percentile)

p75 decimal - 75th percentile

p90 decimal - 90th percentile

p99 decimal - 99th percentile

openrouter: PercentileThroughputCutoffs

Percentile-based throughput cutoffs. All specified cutoffs must be met for an endpoint to be preferred.

Fields

p50? decimal - Minimum p50 throughput (tokens/sec)

p75? decimal - Minimum p75 throughput (tokens/sec)

p90? decimal - Minimum p90 throughput (tokens/sec)

p99? decimal - Minimum p99 throughput (tokens/sec)

openrouter: PerRequestLimits

Per-request token limits

Fields

prompt_tokens decimal - Maximum prompt tokens per request

completion_tokens decimal - Maximum completion tokens per request

openrouter: ProviderPreferences

Provider routing preferences for the request.

Fields

allow_fallbacks? boolean - Whether to allow backup providers to serve requests
- true: (default) when the primary provider (or your custom providers in "order") is unavailable, use the next best provider.
- false: use only the primary/custom provider, and return the upstream error if it's unavailable.

require_parameters? boolean - Whether to filter providers to only those that support the parameters you've provided. If this setting is omitted or set to false, then providers will receive only the parameters they support, and ignore the rest.

data_collection? DataCollection - Data collection setting. If no available model provider meets the requirement, your request will return an error.
- allow: (default) allow providers which store user data non-transiently and may train on it
- deny: use only providers which do not collect user data.

zdr? boolean - Whether to restrict routing to only ZDR (Zero Data Retention) endpoints. When true, only endpoints that do not retain prompts will be used.

enforce_distillable_text? boolean - Whether to restrict routing to only models that allow text distillation. When true, only models where the author has allowed distillation will be used.

'order? (ProviderName|string)[] - An ordered list of provider slugs. The router will attempt to use the first provider in the subset of this list that supports your requested model, and fall back to the next if it is unavailable. If no providers are available, the request will fail with an error message.

only? (ProviderName|string)[] - List of provider slugs to allow. If provided, this list is merged with your account-wide allowed provider settings for this request.

ignore? (ProviderName|string)[] - List of provider slugs to ignore. If provided, this list is merged with your account-wide ignored provider settings for this request.

quantizations? Quantization[] - A list of quantization levels to filter the provider by.

sort? ProviderSort -

max_price? OpenResponsesRequest_provider_max_price - The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion.

preferred_min_throughput? PreferredMinThroughput - Preferred minimum throughput (in tokens per second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints below the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

preferred_max_latency? PreferredMaxLatency - Preferred maximum latency (in seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints above the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

openrouter: ProviderSortConfig

The provider sorting strategy (price, throughput, latency)

Fields

'by? "price"|"throughput"|"latency" - The provider sorting strategy (price, throughput, latency)

partition? "model"|"none" - Partitioning strategy for sorting: "model" (default) groups endpoints by model before sorting (fallback models remain fallbacks), "none" sorts all endpoints together regardless of model.

openrouter: ProvidersResponse

Fields

data record { name string, slug string, privacy_policy_url string, terms_of_service_url string, status_page_url string }[] -

openrouter: PublicEndpoint

Information about a specific model endpoint

Fields

name string -

model_id string - The unique identifier for the model (permaslug)

model_name string -

context_length decimal -

pricing PublicEndpoint_pricing -

provider_name ProviderName -

tag string -

quantization Quantization -

max_completion_tokens decimal -

max_prompt_tokens decimal -

supported_parameters Parameter[] -

status? EndpointStatus -

uptime_last_30m decimal -

supports_implicit_caching boolean -

latency_last_30m PercentileStats - Latency percentiles in milliseconds over the last 30 minutes. Latency measures time to first token. Only visible when authenticated with an API key or cookie; returns null for unauthenticated requests.

throughput_last_30m PercentileStats -

openrouter: PublicEndpoint_pricing

Fields

prompt BigNumberUnion -

completion BigNumberUnion -

request? BigNumberUnion -

image? BigNumberUnion -

image_token? BigNumberUnion -

image_output? BigNumberUnion -

audio? BigNumberUnion -

audio_output? BigNumberUnion -

input_audio_cache? BigNumberUnion -

web_search? BigNumberUnion -

internal_reasoning? BigNumberUnion -

input_cache_read? BigNumberUnion -

input_cache_write? BigNumberUnion -

discount? decimal -

openrouter: PublicPricing

Pricing information for the model

Fields

prompt BigNumberUnion -

completion BigNumberUnion -

request? BigNumberUnion -

image? BigNumberUnion -

image_token? BigNumberUnion -

image_output? BigNumberUnion -

audio? BigNumberUnion -

audio_output? BigNumberUnion -

input_audio_cache? BigNumberUnion -

web_search? BigNumberUnion -

internal_reasoning? BigNumberUnion -

input_cache_read? BigNumberUnion -

input_cache_write? BigNumberUnion -

discount? decimal -

openrouter: ReasoningDetailEncrypted

Reasoning detail encrypted schema

Fields

'type "reasoning.encrypted" -

data string -

id? string -

format? "unknown"|"openai-responses-v1"|"azure-openai-responses-v1"|"xai-responses-v1"|"anthropic-claude-v1"|"google-gemini-v1" -

index? decimal -

openrouter: ReasoningDetailSummary

Reasoning detail summary schema

Fields

'type "reasoning.summary" -

summary string -

id? string -

format? "unknown"|"openai-responses-v1"|"azure-openai-responses-v1"|"xai-responses-v1"|"anthropic-claude-v1"|"google-gemini-v1" -

index? decimal -

openrouter: ReasoningDetailText

Reasoning detail text schema

Fields

'type "reasoning.text" -

text? string -

signature? string -

id? string -

format? "unknown"|"openai-responses-v1"|"azure-openai-responses-v1"|"xai-responses-v1"|"anthropic-claude-v1"|"google-gemini-v1" -

index? decimal -

openrouter: ReasoningSummaryText

Fields

'type "summary_text" -

text string -

openrouter: ReasoningTextContent

Fields

'type "reasoning_text" -

text string -

openrouter: RemoveKeysRequest

Fields

key_hashes RemoveKeyHash[] - Array of API key hashes to unassign from the guardrail

openrouter: RemoveKeysResponse

Fields

unassigned_count decimal - Number of keys successfully unassigned

openrouter: RemoveMembersRequest

Fields

member_user_ids RemoveMemberUserId[] - Array of member user IDs to unassign from the guardrail

openrouter: RemoveMembersResponse

Fields

unassigned_count decimal - Number of members successfully unassigned

openrouter: ResponseFormatJSONObject

JSON object response format

Fields

'type "json_object" -

openrouter: ResponseFormatJSONSchema

JSON Schema response format for structured outputs

Fields

'type "json_schema" -

json_schema JSONSchemaConfig - JSON Schema configuration object

openrouter: ResponseFormatText

Default text response format

Fields

'type "text" -

openrouter: ResponseFormatTextGrammar

Custom grammar response format

Fields

'type "grammar" -

grammar string - Custom grammar for text generation

openrouter: ResponseFormatTextPython

Python code response format

Fields

'type "python" -

openrouter: ResponseInputAudio

Audio input content item

Fields

'type "input_audio" -

input_audio ResponseInputAudio_input_audio -

openrouter: ResponseInputAudio_input_audio

Fields

data string -

format "mp3"|"wav" -

openrouter: ResponseInputFile

File input content item

Fields

'type "input_file" -

file_id? string -

file_data? string -

filename? string -

file_url? string -

openrouter: ResponseInputImage

Image input content item

Fields

'type "input_image" -

detail "auto"|"high"|"low" -

image_url? string -

openrouter: ResponseInputText

Text input content item

Fields

'type "input_text" -

text string -

openrouter: ResponseInputVideo

Video input content item

Fields

'type "input_video" -

video_url string - A base64 data URL or remote URL that resolves to a video file

openrouter: ResponseOutputText

Fields

'type "output_text" -

text string -

annotations? OpenAIResponsesAnnotation[] -

logprobs? record { token string, bytes decimal[], logprob decimal, top_logprobs record { token string, bytes decimal[], logprob decimal }[] }[] -

openrouter: ResponsesErrorField

Error information returned from the API

Fields

code "server_error"|"rate_limit_exceeded"|"invalid_prompt"|"vector_store_timeout"|"invalid_image"|"invalid_image_format"|"invalid_base64_image"|"invalid_image_url"|"image_too_large"|"image_too_small"|"image_parse_error"|"image_content_policy_violation"|"invalid_image_mode"|"image_file_too_large"|"unsupported_image_media_type"|"empty_image_file"|"failed_to_download_image"|"image_file_not_found" -

message string -

openrouter: ResponsesFormatText

Plain text response format

Fields

'type "text" -

openrouter: ResponsesFormatTextJSONSchemaConfig

JSON schema constrained response format

Fields

'type "json_schema" -

name string -

description? string -

strict? boolean -

schema record {} -

openrouter: ResponsesImageGenerationCall

Fields

Fields Included from *OutputItemImageGenerationCall

type "image_generation_call"
id string
result string
status ImageGenerationStatus
anydata...

openrouter: ResponsesOutputItemFileSearchCall

Fields

Fields Included from *OutputItemFileSearchCall

type "file_search_call"
id string
queries string[]
status WebSearchStatus
anydata...

openrouter: ResponsesOutputItemFunctionCall

Fields

Fields Included from *OutputItemFunctionCall

type "function_call"
id string
name string
arguments string
call_id string
status "completed"|"incomplete"|"in_progress"
anydata...

openrouter: ResponsesOutputItemReasoning

An output item containing reasoning

Fields

Fields Included from *OutputItemReasoning

type "reasoning"
id string
content ReasoningTextContent[]
summary ReasoningSummaryText[]
encrypted_content string
status "completed"|"incomplete"|"in_progress"
anydata...

content? ReasoningTextContent[] -

signature? string -

format? "unknown"|"openai-responses-v1"|"azure-openai-responses-v1"|"xai-responses-v1"|"anthropic-claude-v1"|"google-gemini-v1" -

openrouter: ResponsesOutputMessage

An output message item

Fields

Fields Included from *OutputMessage

id string
role "assistant"
type "message"
status "completed"|"incomplete"|"in_progress"
content ResponseOutputText|OpenAIResponsesRefusalContent[]
phase anydata
anydata...

openrouter: ResponsesWebSearchCallOutput

Fields

Fields Included from *OutputItemWebSearchCall

type "web_search_call"
id string
action record { type "search", query string, queries string[], sources record { type "url", url string, anydata... }[], anydata... }|record { type "open_page", url string, anydata... }|record { type "find_in_page", pattern string, url string, anydata... }
status WebSearchStatus
anydata...

openrouter: ResponsesWebSearchUserLocation

User location information for web search

Fields

'type? "approximate" -

city? string -

country? string -

region? string -

timezone? string -

openrouter: ResponseTextConfig

Text output configuration including format and verbosity

Fields

format? ResponseFormatTextConfig - Text response format configuration

verbosity? "high"|"low"|"medium" -

openrouter: SendChatCompletionRequestHeaders

Represents the Headers record for the operation: sendChatCompletionRequest

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: SystemMessage

System message for setting behavior

Fields

role "system" -

content string|ChatMessageContentItemText[] - System message content

name? string - Optional name for the system message

openrouter: ToolDefinitionJson

Tool definition for function calling

Fields

'type "function" -

'function ToolDefinitionJson_function - Function definition for tool calling

cache_control? ChatMessageContentItemCacheControl - Cache control for the content part

openrouter: ToolDefinitionJson_function

Function definition for tool calling

Fields

name string - Function name (a-z, A-Z, 0-9, underscores, dashes, max 64 chars)

description? string - Function description for the model

parameters? record {} - Function parameters as JSON Schema object

strict? boolean - Enable strict schema adherence

openrouter: ToolResponseMessage

Tool response message

Fields

role "tool" -

content string|ChatMessageContentItem[] - Tool response content

tool_call_id string - ID of the assistant message tool call this message responds to

openrouter: TopProviderInfo

Information about the top provider for this model

Fields

context_length? decimal - Context length from the top provider

max_completion_tokens? decimal - Maximum completion tokens from the top provider

is_moderated boolean - Whether the top provider moderates content

openrouter: UpdatedApiKeyInfo

The updated API key information

Fields

hash string - Unique hash identifier for the API key

name string - Name of the API key

label string - Human-readable label for the API key

disabled boolean - Whether the API key is disabled

'limit decimal - Spending limit for the API key in USD

limit_remaining decimal - Remaining spending limit in USD

limit_reset string - Type of limit reset for the API key

include_byok_in_limit boolean - Whether to include external BYOK usage in the credit limit

usage decimal - Total OpenRouter credit usage (in USD) for the API key

usage_daily decimal - OpenRouter credit usage (in USD) for the current UTC day

usage_weekly decimal - OpenRouter credit usage (in USD) for the current UTC week (Monday-Sunday)

usage_monthly decimal - OpenRouter credit usage (in USD) for the current UTC month

byok_usage decimal - Total external BYOK usage (in USD) for the API key

byok_usage_daily decimal - External BYOK usage (in USD) for the current UTC day

byok_usage_weekly decimal - External BYOK usage (in USD) for the current UTC week (Monday-Sunday)

byok_usage_monthly decimal - External BYOK usage (in USD) for current UTC month

created_at string - ISO 8601 timestamp of when the API key was created

updated_at string - ISO 8601 timestamp of when the API key was last updated

expires_at? string - ISO 8601 UTC timestamp when the API key expires, or null if no expiration

openrouter: UpdatedGuardrailData

The updated guardrail

Fields

id string - Unique identifier for the guardrail

name string - Name of the guardrail

description? string - Description of the guardrail

limit_usd? decimal - Spending limit in USD

reset_interval? "daily"|"weekly"|"monthly" - Interval at which the limit resets (daily, weekly, monthly)

allowed_providers? string[] - List of allowed provider IDs

ignored_providers? string[] - List of provider IDs to exclude from routing

allowed_models? string[] - Array of model canonical_slugs (immutable identifiers)

enforce_zdr? boolean - Whether to enforce zero data retention

created_at string - ISO 8601 timestamp of when the guardrail was created

updated_at? string - ISO 8601 timestamp of when the guardrail was last updated

openrouter: UpdateGuardrailHeaders

Represents the Headers record for the operation: updateGuardrail

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: UpdateGuardrailRequest

Fields

name? string - New name for the guardrail

description? string - New description for the guardrail

limit_usd? decimal - New spending limit in USD

reset_interval? "daily"|"weekly"|"monthly" - Interval at which the limit resets (daily, weekly, monthly)

allowed_providers? string[] - New list of allowed provider IDs

ignored_providers? string[] - List of provider IDs to exclude from routing

allowed_models? string[] - Array of model identifiers (slug or canonical_slug accepted)

enforce_zdr? boolean - Whether to enforce zero data retention

openrouter: UpdateGuardrailResponse

Fields

data UpdatedGuardrailData - The updated guardrail

openrouter: UpdateKeyRequest

Fields

name? string - New name for the API key

disabled? boolean - Whether to disable the API key

'limit? decimal - New spending limit for the API key in USD

limit_reset? "daily"|"weekly"|"monthly" - New limit reset type for the API key (daily, weekly, monthly, or null for no reset). Resets happen automatically at midnight UTC, and weeks are Monday through Sunday.

include_byok_in_limit? boolean - Whether to include BYOK usage in the limit

openrouter: UpdateKeyResponse

Fields

data UpdatedApiKeyInfo - The updated API key information

openrouter: UpdateKeysHeaders

Represents the Headers record for the operation: updateKeys

Fields

X\-OpenRouter\-Categories? string - Comma-separated list of app categories (e.g. "cli-agent,cloud-agent"). Used for marketplace rankings.

HTTP\-Referer? string - The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.

X\-OpenRouter\-Title? string - The app display name allows you to customize how your app appears in OpenRouter's dashboard.

openrouter: URLCitation

Fields

'type "url_citation" -

url string -

title string -

start_index decimal -

end_index decimal -

openrouter: UserMessage

User message

Fields

role "user" -

content string|ChatMessageContentItem[] - User message content

name? string - Optional name for the user

openrouter: VideoInput

Video input object

Fields

url string - URL of the video (data: URLs supported)

openrouter: Web3Data

Fields

transfer_intent Web3TransferIntent -

openrouter: Web3TransferIntent

Fields

call_data Web3TransferIntentCallData -

metadata Web3TransferIntentMetadata -

openrouter: Web3TransferIntentCallData

Fields

deadline string -

fee_amount string -

id string -

operator string -

prefix string -

recipient string -

recipient_amount string -

recipient_currency string -

refund_destination string -

signature string -

openrouter: Web3TransferIntentMetadata

Fields

chain_id decimal -

contract_address string -

sender string -

openrouter: WebSearchPreviewToolUserLocation

Fields

'type "approximate" -

city? string -

country? string -

region? string -

timezone? string -

openrouter: ZdrEndpointsResponse

Fields

data PublicEndpoint[] -

Union types

openrouter: ModelGroup

ModelGroup

Tokenizer type used by the model

openrouter: OutputModality

"text"|"image"|"embeddings"|"audio"

OutputModality

openrouter: WebSearchStatus

"completed"|"searching"|"in_progress"|"failed"

WebSearchStatus

openrouter: OpenAIResponsesResponseStatus

OpenAIResponsesResponseStatus

openrouter: ToolCallStatus

"in_progress"|"completed"|"incomplete"

ToolCallStatus

openrouter: PDFParserEngine

"mistral-ocr"|"pdf-text"|"native"

PDFParserEngine

The engine to use for parsing PDF files.

openrouter: Parameter

Parameter

openrouter: ResponsesOutputModality

"text"|"image"

ResponsesOutputModality

openrouter: ChatCompletionFinishReason

"tool_calls"|"stop"|"length"|"content_filter"|"error"

ChatCompletionFinishReason

openrouter: ChatMessageContentItem

ChatMessageContentItem

Content part for chat completion messages

openrouter: ReasoningDetailUnion

ReasoningDetailSummary|ReasoningDetailEncrypted|ReasoningDetailText

ReasoningDetailUnion

Reasoning detail union schema

openrouter: OpenAIResponsesAnnotation

FileCitation|URLCitation|FilePath

OpenAIResponsesAnnotation

openrouter: DataCollection

"deny"|"allow"

DataCollection

Data collection setting. If no available model provider meets the requirement, your request will return an error.

allow: (default) allow providers which store user data non-transiently and may train on it
deny: use only providers which do not collect user data.

openrouter: ResponseFormatTextConfig

ResponsesFormatText|ResponseFormatJSONObject|ResponsesFormatTextJSONSchemaConfig

ResponseFormatTextConfig

Text response format configuration

openrouter: ResponsesOutputItem

ResponsesOutputItem

An output item from the response

openrouter: Quantization

"int4"|"int8"|"fp4"|"fp6"|"fp8"|"fp16"|"bf16"|"fp32"|"unknown"

Quantization

openrouter: ProviderName

ProviderName

openrouter: OpenAIResponsesServiceTier

"auto"|"default"|"flex"|"priority"|"scale"

OpenAIResponsesServiceTier

openrouter: OpenAIResponsesToolChoice

"auto"|"none"|"required"|record { 'type "function" , name string }|record { 'type "web_search_preview_2025_03_11"|"web_search_preview" }

OpenAIResponsesToolChoice

openrouter: ImageGenerationStatus

"in_progress"|"completed"|"generating"|"failed"

ImageGenerationStatus

openrouter: InstructType

InstructType

Instruction format type

openrouter: OpenAIResponsesIncludable

"file_search_call.results"|"message.input_image.image_url"|"computer_call_output.output.image_url"|"reasoning.encrypted_content"|"code_interpreter_call.outputs"

OpenAIResponsesIncludable

openrouter: ToolChoiceOption

"none"|"auto"|"required"|NamedToolChoice

ToolChoiceOption

Tool choice configuration

openrouter: ResponsesSearchContextSize

"low"|"medium"|"high"

ResponsesSearchContextSize

Size of the search context for web search tools

openrouter: ProviderSort

"price"|"throughput"|"latency"

ProviderSort

The provider sorting strategy (price, throughput, latency)

openrouter: InputModality

"text"|"image"|"file"|"audio"|"video"

InputModality

openrouter: OpenAIResponsesTruncation

"auto"|"disabled"

OpenAIResponsesTruncation

openrouter: WebSearchEngine

"native"|"exa"

WebSearchEngine

The search engine to use for web search.

openrouter: EndpointStatus

0|-1|-2|-3|-5|-10

EndpointStatus

openrouter: Message

SystemMessage|UserMessage|DeveloperMessage|AssistantMessage|ToolResponseMessage

Message

Chat completion message with role-based discrimination

openrouter: OpenAIResponsesReasoningEffort

OpenAIResponsesReasoningEffort

openrouter: ReasoningSummaryVerbosity

"auto"|"concise"|"detailed"

ReasoningSummaryVerbosity

Array types

openrouter: ModelsListResponseData

Model[]

ModelsListResponseData

List of available models

openrouter: AssistantMessageReasoningDetails

ReasoningDetailUnion[]

AssistantMessageReasoningDetails

Reasoning details for extended thinking models

openrouter: ModelNames

ModelName[]

ModelNames

Models to use for completion

openrouter: AssistantMessageImages

record { image_url record { url string } }[]

AssistantMessageImages

Generated images from image generation models

Anydata types

openrouter: PreferredMaxLatency

anydata

PreferredMaxLatency

Preferred maximum latency (in seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints above the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

openrouter: OpenAIResponsesInput

anydata

OpenAIResponsesInput

openrouter: PreferredMinThroughput

anydata

PreferredMinThroughput

Preferred minimum throughput (in tokens per second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints below the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold.

openrouter: OpenResponsesInput

anydata

OpenResponsesInput

Input for a response request - can be a string or array of items

String types

openrouter: BigNumberUnion

string

BigNumberUnion

Price per million prompt tokens

openrouter: AssignMemberUserId

string

AssignMemberUserId

openrouter: RemoveMemberUserId

string

RemoveMemberUserId

openrouter: ModelName

string

ModelName

Model to use for completion

openrouter: RemoveKeyHash

string

RemoveKeyHash

openrouter: AssignKeyHash

string

AssignKeyHash

Import

import ballerinax/openrouter;

Other versions

1.0.0

Metadata

Released date: 16 days ago

Version: 1.0.0

License: Apache-2.0

Compatibility

Platform: any

Ballerina version: 2201.12.0

GraalVM compatible: Yes

Pull count

Total: 158

Current verison: 158

Weekly downloads

Source repository

Keywords

Agent

Model

Provider

Vendor/OpenRouter

Area/AI

Type/Connector

Contributors

Dependencies

ballerina/constraint/1.7.0 ballerina/log/2.14.0 ballerina/data.jsondata/1.1.3

Dependents

ballerinax/ai.openrouter/1.0.0

Cookie policy

Delete policy

clients

records

unionTypes

arrayTypes

anyDataTypes

stringTypes

ballerinax/openrouter Ballerina library

Overview

Setup guide

Step 1: Create an OpenRouter Account

Step 2: Generate an API Access Token

Quickstart

Step 1: Import the module

Step 2: Instantiate a new connector

Step 3: Invoke the connector operation

Create a chat completion

Step 4: Run the Ballerina application

Examples

Clients

openrouter: Client

Constructor

post responses

Parameters

Return Type

post messages

Parameters

Return Type

get activity

Parameters

Return Type

post chat/completions

Parameters

Return Type

get credits

Parameters

Return Type

post credits/coinbase

Parameters

Return Type

post embeddings

Parameters

Return Type

get embeddings/models

Parameters

Return Type

get generation

Parameters

Return Type

get models/count

Parameters

Return Type

get models

Parameters

Return Type

get models/user

Parameters

Return Type

get models/[string author]/[string slug]/endpoints

Parameters

Return Type

get endpoints/zdr

Parameters

Return Type

get providers

Parameters

Return Type

get keys

Parameters

Return Type

post keys

Parameters

Return Type

get keys/[string hash]

Parameters

Return Type

delete keys/[string hash]

Parameters

Return Type

patch keys/[string hash]

Parameters