API

Every openrouter.api.* operation is listed below with parameter shapes and return types from the plugin Zod schemas.

New to Corsair? See API access, authentication, and error handling.

Chat Completions

create

chatCompletions.create Generate an AI response via OpenRouter with automatic multi-provider routing, retries, and fallbacks; supports tool calling, structured output, reasoning, and provider/route overrides Risk: write

await corsair.openrouter.api.chatCompletions.create({});

Input

Name	Type	Required	Description
`model`	`string`	Yes	—
`messages`	`object[]`	Yes	—
`stream`	`false`	No	—
`temperature`	`number`	No	—
`topP`	`number`	No	—
`maxTokens`	`number`	No	—
`maxCompletionTokens`	`number`	No	—
`n`	`number`	No	—
`stop`	`string \| string[]`	No	—
`presencePenalty`	`number`	No	—
`frequencyPenalty`	`number`	No	—
`logitBias`	`object`	No	—
`user`	`string`	No	—
`responseFormat`	`object`	No	—
`tools`	`object[]`	No	—
`toolChoice`	`object`	No	—
`reasoning`	`object`	No	—
`transforms`	`string[]`	No	—
`models`	`string[]`	No	—
`route`	`string`	No	—
`provider`	`object`	No	—
`plugins`	`object[]`	No	—

messages full type

(
  {
    role: system,
    content: string
  } | {
    role: assistant,
    content?: string | (
      {
        type: text,
        text: string
      } | {
        type: image_url,
        image_url: {
          url: string,
          detail?: string
        }
      } | {
        type: input_audio,
        input_audio: {
          data: string,
          format: string
        }
      }
    )[] | null,
    tool_calls?: {
      id: string,
      type: function,
      function: {
        name: string,
        arguments: string
      }
    }[]
  } | {
    role: user,
    content: string | (
      {
        type: text,
        text: string
      } | {
        type: image_url,
        image_url: {
          url: string,
          detail?: string
        }
      } | {
        type: input_audio,
        input_audio: {
          data: string,
          format: string
        }
      }
    )[]
  } | {
    role: tool,
    tool_call_id: string,
    content: string | (
      {
        type: text,
        text: string
      } | {
        type: image_url,
        image_url: {
          url: string,
          detail?: string
        }
      } | {
        type: input_audio,
        input_audio: {
          data: string,
          format: string
        }
      }
    )[]
  }
)[]

logitBias full type

{
}

responseFormat full type

{
  type: text | json_object | json_schema,
  json_schema?: {
    name: string,
    strict?: boolean,
    schema?: {
    }
  }
}

tools full type

{
  type: function,
  function: {
    name: string,
    description?: string,
    parameters?: {
    }
  }
}[]

toolChoice full type

string | {
  type: function,
  function: {
    name: string
  }
}

reasoning full type

{
  effort?: none | minimal | low | medium | high | xhigh | max,
  summary?: auto | concise | detailed | null
}

provider full type

{
  order?: string[],
  allow_fallbacks?: boolean,
  ignore?: string[],
  require_parameters?: boolean,
  data_collection?: string,
  zdr?: boolean
}

plugins full type

{
  name: string,
  max_tokens?: number,
  num_images_per_prompt?: number,
  image_format?: string,
  num_prompts?: number
}[]

Output

Name	Type	Required	Description
`id`	`string`	Yes	—
`object`	`chat.completion`	Yes	—
`created`	`number`	Yes	—
`model`	`string`	Yes	—
`choices`	`object[]`	Yes	—
`usage`	`object`	No	—
`provider`	`string`	No	—
`models`	`string[]`	No	—
`native_tool_calls`	`any[]`	No	—

choices full type

{
  index: number,
  message: {
    role: assistant,
    content?: string | null,
    tool_calls?: {
      id: string,
      type: function,
      function: {
        name: string,
        arguments: string
      }
    }[],
    reasoning?: string | null,
    reasoning_details?: {
    }[]
  },
  finish_reason?: string | null
}[]

usage full type

{
  prompt_tokens: number,
  completion_tokens: number,
  total_tokens: number,
  prompt_tokens_details?: {
    cached_tokens?: number,
    cache_write_tokens?: number,
    audio_tokens?: number,
    video_tokens?: number
  },
  completion_tokens_details?: {
    reasoning_tokens?: number,
    audio_tokens?: number,
    image_tokens?: number
  },
  cost?: number,
  is_byok?: boolean
}

Credits

list

credits.list Get the account credit balance and usage with a management API key Risk: read

await corsair.openrouter.api.credits.list({});

Input: empty object Output

Name	Type	Required	Description
`data`	`object`	Yes	—

data full type

{
  total_credits: number,
  total_usage: number,
  limit_reached?: boolean,
  prepaid?: number,
  billed_prepaid?: number,
  soft_limit?: number,
  pending_balance?: number
}

Embeddings

create

embeddings.create Generate vector embeddings for one or more input strings using a supported embedding model Risk: write

await corsair.openrouter.api.embeddings.create({});

Input

Name	Type	Required	Description
`model`	`string`	Yes	—
`input`	`object[]`	Yes	—
`encodingFormat`	`float \| base64`	No	—
`dimensions`	`number`	No	—
`user`	`string`	No	—
`inputType`	`string`	No	—
`provider`	`object`	No	—

input full type

string | string[] | number[] | number[][] | {
}[]

provider full type

{
  order?: string[],
  allow_fallbacks?: boolean,
  ignore?: string[],
  require_parameters?: boolean,
  data_collection?: string,
  zdr?: boolean
}

Output

Name	Type	Required	Description
`id`	`string`	No	—
`object`	`list`	Yes	—
`data`	`object[]`	Yes	—
`model`	`string`	Yes	—
`usage`	`object`	No	—

data full type

{
  index?: number,
  object: embedding,
  embedding: number[] | string
}[]

usage full type

{
  prompt_tokens: number,
  total_tokens: number
}

Generations

get

generations.get Fetch request and usage metadata for a previous generation by its ID Risk: read

await corsair.openrouter.api.generations.get({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—

Output

Name	Type	Required	Description
`data`	`object`	Yes	—

data full type

{
  id: string,
  model?: string,
  provider?: string,
  provider_name?: string | null,
  api_type?: string | null,
  created_at?: string,
  streamed?: boolean | null,
  finish_reason?: string | null,
  total_cost?: number | null,
  prompt_tokens?: number,
  completion_tokens?: number,
  total_tokens?: number,
  usage?: number | {
  },
  tokens_prompt?: number | null,
  tokens_completion?: number | null,
  provider_responses?: any[] | null,
  provider_response?: {
  }
}

Key

get

key.get Get metadata about the current API key, including usage, limits, and rate limits Risk: read

await corsair.openrouter.api.key.get({});

Input: empty object Output

Name	Type	Required	Description
`data`	`object`	Yes	—

data full type

{
  label?: string,
  usage: number,
  usage_daily?: number,
  usage_weekly?: number,
  usage_monthly?: number,
  byok_usage?: number,
  byok_usage_daily?: number,
  byok_usage_weekly?: number,
  byok_usage_monthly?: number,
  limit?: number | null,
  limit_reset?: string | null,
  limit_remaining?: number | null,
  include_byok_in_limit?: boolean,
  creator_user_id?: string | null,
  is_free_tier?: boolean,
  is_management_key?: boolean,
  is_provisioning_key?: boolean,
  rate_limit?: {
    requests: number,
    interval: string
  },
  expires_at?: string | null,
  created_at?: string
}

Messages

create

messages.create Create a message via OpenRouter’s Anthropic Messages API, with support for system prompts and multi-part content Risk: write

await corsair.openrouter.api.messages.create({});

Input

Name	Type	Required	Description
`model`	`string`	Yes	—
`maxTokens`	`number`	No	—
`messages`	`object[]`	Yes	—
`system`	`string`	No	—
`temperature`	`number`	No	—
`topP`	`number`	No	—
`stopSequences`	`string[]`	No	—
`tools`	`object[]`	No	—
`toolChoice`	`object`	No	—
`thinking`	`object`	No	—

messages full type

{
  role: user | assistant,
  content: string | (
    {
      type: text,
      text: string
    } | {
      type: image,
      source: {
        type: base64,
        media_type: string,
        data: string
      } | {
        type: url,
        url: string
      }
    } | {
      type: document,
      source: {
        type: base64,
        media_type: string,
        data: string
      } | {
        type: url,
        url: string
      } | {
        type: text,
        media_type: text/plain,
        data: string
      } | {
        type: content,
        content: string | (
          {
            type: text,
            text: string
          } | {
            type: image,
            source: {
              type: base64,
              media_type: string,
              data: string
            } | {
              type: url,
              url: string
            }
          }
        )[]
      } | {
        type: file,
        file_id: string
      },
      title?: string | null,
      context?: string | null,
      citations?: {
        enabled?: boolean
      } | null
    } | {
      type: tool_use,
      id: string,
      name: string,
      input: {
      }
    } | {
      type: tool_result,
      tool_use_id: string,
      content: string | any[],
      is_error?: boolean
    } | {
      type: thinking,
      thinking: string,
      signature?: string
    } | {
      type: redacted_thinking,
      data: string
    }
  )[]
}[]

tools full type

{
  name: string,
  description?: string,
  input_schema: {
  }
}[]

toolChoice full type

{
  type: tool,
  name: string
} | {
  type: auto | any | none
}

thinking full type

{
  type: enabled,
  budget_tokens: number
} | {
  type: disabled
}

Output

Name	Type	Required	Description
`id`	`string`	Yes	—
`type`	`message`	Yes	—
`role`	`assistant`	Yes	—
`model`	`string`	Yes	—
`stop_reason`	`string`	No	—
`content`	`object[]`	Yes	—
`usage`	`object`	Yes	—
`provider`	`string`	No	—

content full type

(
  {
    type: text,
    text: string,
    citations?: any[] | null
  } | {
    type: thinking,
    thinking: string,
    signature?: string
  } | {
    type: redacted_thinking,
    data: string
  } | {
    type: tool_use,
    id: string,
    name: string,
    input: {
    }
  }
)[]

usage full type

{
  input_tokens: number,
  output_tokens: number,
  cache_read_input_tokens?: number | null,
  cache_creation_input_tokens?: number | null,
  output_tokens_details?: {
    thinking_tokens?: number
  } | null
}

Model Endpoints

list

modelEndpoints.list List the individual endpoints serving a model, with per-provider pricing, latency, and throughput Risk: read

await corsair.openrouter.api.modelEndpoints.list({});

Input

Name	Type	Required	Description
`author`	`string`	Yes	—
`slug`	`string`	Yes	—

Output

Name	Type	Required	Description
`data`	`object`	Yes	—

data full type

{
  id: string,
  name: string,
  created?: number,
  description?: string,
  architecture?: {
  },
  endpoints: {
    name: string,
    model_id?: string,
    model_name?: string,
    provider_name: string,
    tag?: string,
    context_length?: number,
    max_completion_tokens?: number | null,
    max_prompt_tokens?: number | null,
    quantization?: string | null,
    pricing?: {
    },
    supported_parameters?: string[],
    status?: number,
    uptime_last_30m?: number | null,
    uptime_last_5m?: number | null,
    uptime_last_1d?: number | null,
    supports_implicit_caching?: boolean,
    supports_voice_cloning?: boolean,
    latency_last_30m?: {
      p50?: number,
      p75?: number,
      p90?: number,
      p99?: number
    } | null,
    throughput_last_30m?: {
      p50?: number,
      p75?: number,
      p90?: number,
      p99?: number
    } | null
  }[]
}

Models

count

models.count Get the total count of models available on OpenRouter Risk: read

await corsair.openrouter.api.models.count({});

Input: empty object Output

Name	Type	Required	Description
`data`	`object`	Yes	—

data full type

{
  count: number
}

list

models.list List all models available on OpenRouter, including pricing, context length, and supported parameters Risk: read

await corsair.openrouter.api.models.list({});

Input

Name	Type	Required	Description
`offset`	`number`	No	—
`limit`	`number`	No	—

Output

Name	Type	Required	Description
`data`	`object[]`	Yes	—
`links`	`object`	No	—
`total_count`	`number`	No	—

data full type

{
  id: string,
  name?: string,
  created?: number,
  description?: string,
  context_length?: number,
  pricing?: {
  },
  architecture?: {
    modality?: string,
    input_modalities?: string[],
    output_modalities?: string[],
    tokenizer?: string,
    instruct_type?: string | null
  },
  top_provider?: {
    context_length?: number | null,
    max_completion_tokens?: number | null,
    is_moderated?: boolean
  },
  per_request_limits?: {
    prompt_tokens?: string | null,
    completion_tokens?: string | null
  } | null,
  supported_parameters?: string[]
}[]

links full type

{
  next?: string | null
}

listEmbeddings

models.listEmbeddings List all embedding models available on OpenRouter Risk: read

await corsair.openrouter.api.models.listEmbeddings({});

Input

Name	Type	Required	Description
`offset`	`number`	No	—
`limit`	`number`	No	—

Output

Name	Type	Required	Description
`data`	`object[]`	Yes	—
`links`	`object`	No	—
`total_count`	`number`	No	—

data full type

{
  id: string,
  name?: string,
  created?: number,
  description?: string,
  context_length?: number,
  pricing?: {
  },
  architecture?: {
    modality?: string,
    input_modalities?: string[],
    output_modalities?: string[],
    tokenizer?: string,
    instruct_type?: string | null
  },
  top_provider?: {
    context_length?: number | null,
    max_completion_tokens?: number | null,
    is_moderated?: boolean
  },
  per_request_limits?: {
    prompt_tokens?: string | null,
    completion_tokens?: string | null
  } | null,
  supported_parameters?: string[]
}[]

links full type

{
  next?: string | null
}

listUser

models.listUser List models filtered by the authenticated user’s provider preferences, privacy settings, and guardrails Risk: read

await corsair.openrouter.api.models.listUser({});

Input

Name	Type	Required	Description
`offset`	`number`	No	—
`limit`	`number`	No	—

Output

Name	Type	Required	Description
`data`	`object[]`	Yes	—
`links`	`object`	No	—
`total_count`	`number`	No	—

data full type

{
  id: string,
  name?: string,
  created?: number,
  description?: string,
  context_length?: number,
  pricing?: {
  },
  architecture?: {
    modality?: string,
    input_modalities?: string[],
    output_modalities?: string[],
    tokenizer?: string,
    instruct_type?: string | null
  },
  top_provider?: {
    context_length?: number | null,
    max_completion_tokens?: number | null,
    is_moderated?: boolean
  },
  per_request_limits?: {
    prompt_tokens?: string | null,
    completion_tokens?: string | null
  } | null,
  supported_parameters?: string[]
}[]

links full type

{
  next?: string | null
}

Providers

list

providers.list List the providers available on OpenRouter with their privacy policies and data-center regions Risk: read

await corsair.openrouter.api.providers.list({});

Input: empty object Output

Name	Type	Required	Description
`data`	`object[]`	Yes	—

data full type

{
  name: string,
  slug: string,
  privacy_policy_url?: string | null,
  terms_of_service_url?: string | null,
  status_page_url?: string | null,
  headquarters?: string | null,
  datacenters?: string[] | null
}[]

Zdr

list

zdr.list List the Zero-Data Residency (ZDR) endpoint specification for the account Risk: read

await corsair.openrouter.api.zdr.list({});

Input: empty object Output

Name	Type	Required	Description
`data`	`object[]`	Yes	—

data full type

{
  name: string,
  model_id?: string,
  model_name?: string,
  provider_name: string,
  tag?: string,
  context_length?: number,
  max_completion_tokens?: number | null,
  max_prompt_tokens?: number | null,
  quantization?: string | null,
  pricing?: {
  },
  supported_parameters?: string[],
  status?: number,
  uptime_last_30m?: number | null,
  uptime_last_5m?: number | null,
  uptime_last_1d?: number | null,
  supports_implicit_caching?: boolean,
  supports_voice_cloning?: boolean,
  latency_last_30m?: {
    p50?: number,
    p75?: number,
    p90?: number,
    p99?: number
  } | null,
  throughput_last_30m?: {
    p50?: number,
    p75?: number,
    p90?: number,
    p99?: number
  } | null
}[]

Overview

Plugins

Chat Completions

create

Credits

list

Embeddings

create

Generations

get

Key

get

Messages

create

Model Endpoints

list

Models

count

list

listEmbeddings

listUser

Providers

list

Zdr

list

​Chat Completions

​create

​Credits

​list

​Embeddings

​create

​Generations

​get

​Key

​get

​Messages

​create

​Model Endpoints

​list

​Models

​count

​list

​listEmbeddings

​listUser

​Providers

​list

​Zdr

​list

Chat Completions

create

Credits

list

Embeddings

create

Generations

get

Key

get

Messages

create

Model Endpoints

list

Models

count

list

listEmbeddings

listUser

Providers

list

Zdr

list