OpenLimits

https://openlimits.app/docs
Admin
OpenLimits is a drop-in proxy for the Anthropic and OpenAI APIs. It allows clients to use a single...
Tokens:8,012
Snippets:57
Trust Score:4.2
Update:3 months ago
Show doc for...
Tokens:
Raw
### Conductor Environment Variable Setup

Source: https://openlimits.app/docs

Set these environment variables to force Conductor to use the Openlimits API if Claude Code CLI is not installed.

```shell
ANTHROPIC_BASE_URL=https://openlimits.app
ANTHROPIC_API_KEY=your-key-here
```

--------------------------------

### POST /v1/chat/completions Response Example

Source: https://openlimits.app/docs

This is an example of a successful JSON response from the /v1/chat/completions endpoint, following the OpenAI chat completion format. It includes details about the conversation, model used, and token usage.

```json
{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "created": 1700000000,
  "model": "anthropic/claude-opus-4.8",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Hello! How can I help?" },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 8,
    "total_tokens": 28
  }
}
```

--------------------------------

### Python Anthropic SDK Example

Source: https://openlimits.app/docs

Use the Anthropic Python SDK to interact with the Openlimits API. Ensure you have the SDK installed and your API key and base URL are configured.

```python
import anthropic

client = anthropic.Anthropic(
    api_key="your-key-here",
    base_url="https://openlimits.app",
)

message = client.messages.create(
    model="anthropic/claude-opus-4.8",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
print(message.content[0].text)
```

--------------------------------

### List Available Models

Source: https://openlimits.app/docs

Use this GET request to retrieve a list of all available models. Include your API key in the 'x-api-key' header.

```http
GET /v1/models
x-api-key: YOUR_KEY

```

```json
{
  "object": "list",
  "data": [
    { "id": "anthropic/claude-opus-4.8", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/claude-opus-4.7", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/claude-sonnet-5", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/claude-haiku-4.5", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/fable-5", "object": "model", "owned_by": "anthropic" },
    { "id": "openai/gpt-5.6-sol", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.6-terra", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.6-luna", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.5", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.4", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.4-mini", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.3-codex-spark", "object": "model", "owned_by": "openai" },
    { "id": "z-ai/glm-5.2", "object": "model", "owned_by": "zai" },
    { "id": "z-ai/glm-5.1", "object": "model", "owned_by": "zai" },
    { "id": "z-ai/glm-5-turbo", "object": "model", "owned_by": "zai" },
    { "id": "minimax/minimax-m3", "object": "model", "owned_by": "minimax" },
    { "id": "deepseek/deepseek-v4-pro", "object": "model", "owned_by": "deepseek" },
    { "id": "deepseek/deepseek-v4-flash", "object": "model", "owned_by": "deepseek" }
  ]
}

```

--------------------------------

### Python OpenAI SDK Example

Source: https://openlimits.app/docs

Use the OpenAI Python SDK to interact with the Openlimits API. Configure the client with your API key and the Openlimits base URL.

```python
from openai import OpenAI

client = OpenAI(
    api_key="your-key-here",
    base_url="https://openlimits.app/v1",
)

response = client.chat.completions.create(
    model="anthropic/claude-opus-4.8",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)
```

--------------------------------

### Set Environment Variables for Anthropic

Source: https://openlimits.app/docs

Configure your environment by setting the ANTHROPIC_BASE_URL and ANTHROPIC_API_KEY. This is the only setup required for Anthropic-compatible clients to route through OpenLimits.

```bash
export ANTHROPIC_BASE_URL=https://openlimits.app
export ANTHROPIC_API_KEY=your-key-here
```

--------------------------------

### POST /v1/responses Request Example

Source: https://openlimits.app/docs

This snippet shows how to make a POST request to the /v1/responses endpoint to get a streamed response from an OpenAI or Fable model. Ensure you have a valid API key and specify the desired model and input.

```http
POST /v1/responses
Content-Type: application/json
Authorization: Bearer YOUR_KEY

{
  "model": "openai/gpt-5.5",
  "instructions": "You are a helpful assistant.",
  "input": "Explain recursion in one sentence.",
  "stream": true,
  "store": false
}
```

--------------------------------

### POST /v1/messages Request Example

Source: https://openlimits.app/docs

This example demonstrates a direct proxy request to the Anthropic Messages API using the /v1/messages endpoint. The request format is identical to Anthropic's.

```http
POST /v1/messages
Content-Type: application/json
x-api-key: YOUR_KEY

{
  "model": "anthropic/claude-opus-4.8",
  "max_tokens": 1024,
  "messages": [
    { "role": "user", "content": "Explain recursion in one sentence." }
  ],
  "stream": false
}
```

--------------------------------

### cURL Request Example

Source: https://openlimits.app/docs

Make a direct API request using cURL to the Openlimits messages endpoint. Ensure you include the correct API key and content type headers.

```shell
curl https://openlimits.app/v1/messages \
  -H "x-api-key: your-key-here" \
  -H "content-type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.8",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello"}]
  }'
```

--------------------------------

### Automated Claude Code CLI Setup

Source: https://openlimits.app/docs

Run this script to automatically configure Claude Code and Codex CLI. Alternatively, manually add settings to ~/.claude/settings.json.

```shell
curl -fsSL "https://openlimits.app/setup?token=your-key-here" | bash
```

```json
{
  "env": {
    "ANTHROPIC_BASE_URL": "https://openlimits.app",
    "ANTHROPIC_AUTH_TOKEN": "your-key-here"
  }
}
```

--------------------------------

### Anthropic Error Format Example

Source: https://openlimits.app/docs

This is an example of the error response format used by Anthropic models.

```json
{
  "type": "error",
  "error": {
    "type": "authentication_error",
    "message": "Invalid or disabled authentication token"
  }
}
```

--------------------------------

### POST /v1/messages Response Example

Source: https://openlimits.app/docs

A typical response from the /v1/messages endpoint, mirroring the Anthropic Messages API response format.

```json
{
  "id": "msg_...",
  "type": "message",
  "role": "assistant",
  "content": [
    { "type": "text", "text": "Recursion is when a function calls itself..." }
  ],
  "model": "anthropic/claude-opus-4.8",
  "usage": {
    "input_tokens": 14,
    "output_tokens": 32,
    "cache_read_input_tokens": 0,
    "cache_creation_input_tokens": 0
  }
}
```

--------------------------------

### TypeScript Anthropic SDK Example

Source: https://openlimits.app/docs

Use the Anthropic TypeScript SDK to interact with the Openlimits API. Set your API key and the base URL for the client.

```typescript
import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "your-key-here",
  baseURL: "https://openlimits.app",
});

const message = await client.messages.create({
  model: "anthropic/claude-opus-4.8",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello" }],
});
console.log(message.content[0].text)
```

--------------------------------

### POST /v1/chat/completions Request Example

Source: https://openlimits.app/docs

This snippet demonstrates a POST request to the /v1/chat/completions endpoint, which is compatible with OpenAI SDK clients. It translates requests to the Anthropic API on the fly. Include your API key, model, messages, and desired stream setting.

```http
POST /v1/chat/completions
Content-Type: application/json
Authorization: Bearer YOUR_KEY

{
  "model": "anthropic/claude-opus-4.8",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "Hello" }
  ],
  "max_tokens": 1024,
  "stream": false
}
```

--------------------------------

### GET /v1/models

Source: https://openlimits.app/docs

Returns a list of all available models in OpenAI-compatible format. This endpoint is useful for discovering which models are available for use through the API.

```APIDOC
## GET /v1/models

### Description
Returns a list of all available models in OpenAI-compatible format.

### Method
GET

### Endpoint
/v1/models

### Headers
- **x-api-key** (string) - Required - Your API key for authentication.

### Response
#### Success Response (200)
- **object** (string) - The type of the response object, expected to be "list".
- **data** (array) - An array of model objects.
  - **id** (string) - The unique identifier for the model.
  - **object** (string) - The type of the object, expected to be "model".
  - **owned_by** (string) - The organization that owns the model.

### Response Example
```json
{
  "object": "list",
  "data": [
    { "id": "anthropic/claude-opus-4.8", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/claude-opus-4.7", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/claude-sonnet-5", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/claude-haiku-4.5", "object": "model", "owned_by": "anthropic" },
    { "id": "anthropic/fable-5", "object": "model", "owned_by": "anthropic" },
    { "id": "openai/gpt-5.6-sol", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.6-terra", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.6-luna", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.5", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.4", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.4-mini", "object": "model", "owned_by": "openai" },
    { "id": "openai/gpt-5.3-codex-spark", "object": "model", "owned_by": "openai" },
    { "id": "z-ai/glm-5.2", "object": "model", "owned_by": "zai" },
    { "id": "z-ai/glm-5.1", "object": "model", "owned_by": "zai" },
    { "id": "z-ai/glm-5-turbo", "object": "model", "owned_by": "zai" },
    { "id": "minimax/minimax-m3", "object": "model", "owned_by": "minimax" },
    { "id": "deepseek/deepseek-v4-pro", "object": "model", "owned_by": "deepseek" },
    { "id": "deepseek/deepseek-v4-flash", "object": "model", "owned_by": "deepseek" }
  ]
}
```
```

--------------------------------

### POST /v1/messages Streaming Request Example

Source: https://openlimits.app/docs

To enable streaming responses from the /v1/messages endpoint, set the 'stream' parameter to true. The event format follows the Anthropic streaming specification.

```json
{
  "model": "anthropic/claude-opus-4.8",
  "max_tokens": 1024,
  "messages": [...],
  "stream": true
}
```

--------------------------------

### Configure OpenAI Clients with OpenLimits

Source: https://openlimits.app/docs

Set environment variables for OpenAI-compatible clients like Codex CLI/Desktop and OpenAI SDKs to use OpenLimits. The base URL includes '/v1'.

```bash
# For Codex / OpenAI clients
export OPENAI_BASE_URL=https://openlimits.app/v1
export OPENAI_API_KEY=your-key-here
```

--------------------------------

### Configure Anthropic Clients with OpenLimits

Source: https://openlimits.app/docs

Set environment variables for Anthropic-compatible clients like Claude Code CLI and Conductor to use OpenLimits. Note the base URL does not include '/v1'.

```bash
# For Claude Code / Anthropic clients
export ANTHROPIC_BASE_URL=https://openlimits.app
export ANTHROPIC_API_KEY=your-key-here
```

--------------------------------

### Configure Codex CLI with OpenLimits

Source: https://openlimits.app/docs

Set up your Codex CLI to use OpenLimits as a model provider by editing the config.toml file. Ensure your API key is set as an environment variable.

```toml
# ~/.codex/config.toml
model = "openai/gpt-5.5"
review_model = "openai/gpt-5.5"
model_provider = "openlimits"

[model_providers.openlimits]
name = "OpenLimits"
base_url = "https://openlimits.app/v1"
env_key = "OPENAI_API_KEY"
wire_api = "responses"
```

--------------------------------

### Configure Codex CLI for GPT-5.6 Sol

Source: https://openlimits.app/docs

To use GPT-5.6 Sol, update the top-level model fields in your Codex CLI config.toml. The existing OpenLimits provider configuration should remain.

```toml
model = "openai/gpt-5.6-sol"
review_model = "openai/gpt-5.6-sol"
model_provider = "openlimits"
```

--------------------------------

### POST /v1/chat/completions

Source: https://openlimits.app/docs

An OpenAI-compatible endpoint that translates requests to the Anthropic API on the fly. Use this with any OpenAI SDK client.

```APIDOC
## POST /v1/chat/completions

### Description
OpenAI-compatible endpoint. Send requests in the OpenAI format and we translate to Anthropic on the fly. Use this with any OpenAI SDK client.

### Method
POST

### Endpoint
/v1/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model to use (e.g., "anthropic/claude-opus-4.8").
- **messages** (array) - Required - An array of message objects, each with a `role` and `content`.
  - **role** (string) - Required - The role of the message sender (e.g., "system", "user", "assistant").
  - **content** (string) - Required - The content of the message.
- **max_tokens** (integer) - Optional - The maximum number of tokens to generate.
- **stream** (boolean) - Optional - Whether to stream the response. Defaults to false.

### Request Example
```json
{
  "model": "anthropic/claude-opus-4.8",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "Hello" }
  ],
  "max_tokens": 1024,
  "stream": false
}
```

### Response
#### Success Response
Returns a JSON object containing the chat completion details.

#### Response Example
```json
{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "created": 1700000000,
  "model": "anthropic/claude-opus-4.8",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Hello! How can I help?" },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 8,
    "total_tokens": 28
  }
}
```

### Model Routing
Claude model names are translated to the Anthropic Messages API format automatically. GPT models use OpenAI provider routes. Fable keeps its public Anthropic-compatible model ID while routing through OpenAI providers as a custom `gpt-5.5`-backed model.
Model| Routes To  
---|---
`anthropic/claude-opus-4.8`, `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-5`, `anthropic/claude-haiku-4.5`| Anthropic-compatible provider pool  
`anthropic/fable-5`| OpenAI provider pool as upstream `gpt-5.5`, with Anthropic-compatible responses on `/v1/messages`  
`minimax/minimax-m3`| MiniMax provider pool  
`z-ai/glm-5.2`, `z-ai/glm-5.1`, `z-ai/glm-5-turbo`, `deepseek/deepseek-v4-pro`, `deepseek/deepseek-v4-flash`| MiniMax fallback provider pool
```

--------------------------------

### Set OpenAI API Key for Codex CLI

Source: https://openlimits.app/docs

Export your OpenAI API key to your shell profile to be used by Codex CLI when connecting to OpenLimits.

```bash
export OPENAI_API_KEY=your-key-here
```

--------------------------------

### POST /v1/responses

Source: https://openlimits.app/docs

Proxies requests to the OpenAI Responses API for GPT and Fable models. This endpoint is used by Codex CLI and Codex Desktop.

```APIDOC
## POST /v1/responses

### Description
Proxies to the **OpenAI Responses API** for GPT and Fable models. This is the endpoint used by **Codex CLI** and **Codex Desktop** when configured with OpenLimits.

### Method
POST

### Endpoint
/v1/responses

### Parameters
#### Request Body
- **model** (string) - Required - The model to use (e.g., "openai/gpt-5.5").
- **instructions** (string) - Optional - Instructions for the model.
- **input** (string) - Required - The input prompt for the model.
- **stream** (boolean) - Required - Always true, as streaming is enabled by default.
- **store** (boolean) - Required - Always false, as required by the upstream provider.

### Request Example
```json
{
  "model": "openai/gpt-5.5",
  "instructions": "You are a helpful assistant.",
  "input": "Explain recursion in one sentence.",
  "stream": true,
  "store": false
}
```

### Response
Returns a server-sent event (SSE) stream. Streaming is always enabled for this endpoint. The event format follows the OpenAI Responses API spec.

### Supported Models
Use OpenRouter-style OpenAI model IDs, or `anthropic/fable-5` for Fable:
  * `anthropic/fable-5`
  * `openai/gpt-5.6-sol`
  * `openai/gpt-5.6-terra`
  * `openai/gpt-5.6-luna`
  * `openai/gpt-5.5`
  * `openai/gpt-5.4`
  * `openai/gpt-5.4-mini`
  * `openai/gpt-5.3-codex-spark`
```

--------------------------------

### POST /v1/messages Effort Levels Configuration

Source: https://openlimits.app/docs

Control the trade-off between quality and speed for model responses using the 'effort' parameter within 'output_config'. Valid values include 'low', 'medium', 'high', 'extra', 'max', and 'ultracode'.

```json
{
  "model": "anthropic/claude-opus-4.8",
  "max_tokens": 1024,
  "messages": [...],
  "output_config": { "effort": "low" }
}
```

--------------------------------

### GPT (OpenAI) Endpoints

Source: https://openlimits.app/docs

Endpoints for interacting with GPT models from OpenAI. Supports chat completions and responses.

```APIDOC
## GPT (OpenAI)

### Description
Access GPT models via the `/v1/chat/completions` and `/v1/responses` endpoints.

### Models
- `openai/gpt-5.6-sol` (GPT-5.6 Sol)
- `openai/gpt-5.6-terra` (GPT-5.6 Terra)
- `openai/gpt-5.6-luna` (GPT-5.6 Luna)
- `openai/gpt-5.5` (GPT-5.5)
- `openai/gpt-5.4` (GPT-5.4)
- `openai/gpt-5.4-mini` (GPT-5.4 Mini)
- `openai/gpt-5.3-codex-spark` (GPT-5.3 Codex Spark)

### Endpoints
- `/v1/chat/completions`
- `/v1/responses`
```

--------------------------------

### Use OpenLimits Model in Codex CLI

Source: https://openlimits.app/docs

Specify the OpenLimits-compatible model ID when running commands in Codex CLI.

```bash
codex --model openai/gpt-5.5
```

--------------------------------

### Authenticate API Requests

Source: https://openlimits.app/docs

API requests to OpenLimits require authentication using either the 'x-api-key' header (Anthropic style) or the 'Authorization: Bearer' header (OpenAI style).

```curl
curl https://openlimits.app/v1/messages \
  -H "x-api-key: YOUR_KEY" \
  -H "content-type: application/json" \
  -d '{"model":"anthropic/claude-opus-4.8","max_tokens":1024,"messages":[{"role":"user","content":"Hello"}]}'
```

--------------------------------

### Claude (Anthropic) Endpoints

Source: https://openlimits.app/docs

Endpoints for interacting with Claude models from Anthropic. Supports both messages and chat completions.

```APIDOC
## Claude (Anthropic)

### Description
Access Claude models via the `/v1/messages` and `/v1/chat/completions` endpoints. Fable 5 also supports `/v1/responses`.

### Models
- `anthropic/claude-opus-4.8` (Opus 4.8)
- `anthropic/claude-opus-4.7` (Opus 4.7)
- `anthropic/claude-sonnet-5` (Sonnet 5)
- `anthropic/claude-haiku-4.5` (Haiku 4.5)
- `anthropic/fable-5` (Fable 5)

### Endpoints
- `/v1/messages`
- `/v1/chat/completions`
- `/v1/responses` (for Fable 5)
```

--------------------------------

### Open Model Groups Endpoints

Source: https://openlimits.app/docs

Endpoints for accessing open model groups using OpenRouter-style IDs. Available on plans with open model access.

```APIDOC
## Open Model Groups

### Description
Access various open models using OpenRouter-style IDs. Available on plans that include open model access.

### Models
- `z-ai/glm-5.2` (GLM 5.2)
- `z-ai/glm-5.1` (GLM 5.1)
- `z-ai/glm-5-turbo` (GLM 5 Turbo)
- `minimax/minimax-m3` (MiniMax M3)
- `deepseek/deepseek-v4-pro` (DeepSeek V4 Pro)
- `deepseek/deepseek-v4-flash` (DeepSeek V4 Flash)

### Endpoints
- `/v1/chat/completions`
- `/v1/responses`
```

--------------------------------

### Error Handling

Source: https://openlimits.app/docs

Details on the error formats and status codes used by the API, following the Anthropic error format.

```APIDOC
## Errors

Errors follow the Anthropic error format:

```json
{
  "type": "error",
  "error": {
    "type": "authentication_error",
    "message": "Invalid or disabled authentication token"
  }
}
```

### Status Codes
- `400` (invalid_request_error): Invalid model or malformed request
- `401` (authentication_error): Missing or invalid API key
- `403` (permission_error): Spend limit exceeded or token expired
- `429` (rate_limit_error): All providers temporarily at capacity
```

--------------------------------

### POST /v1/messages

Source: https://openlimits.app/docs

Proxies requests directly to the Anthropic Messages API. Supports standard Anthropic request and response formats, streaming, effort level control, and extended thinking.

```APIDOC
## POST /v1/messages

### Description
Proxies directly to the Anthropic Messages API. The request and response formats are identical — see Anthropic's docs for the full schema.

### Method
POST

### Endpoint
/v1/messages

### Request Body
- **model** (string) - Required - The model to use for generation.
- **max_tokens** (integer) - Required - The maximum number of tokens to generate.
- **messages** (array) - Required - An array of message objects representing the conversation history.
  - **role** (string) - Required - The role of the message sender (e.g., "user", "assistant").
  - **content** (string) - Required - The content of the message.
- **stream** (boolean) - Optional - Whether to stream the response.
- **output_config** (object) - Optional - Configuration for the output.
  - **effort** (string) - Optional - Controls quality vs speed. Valid values: `low`, `medium`, `high`, `extra`, `max`, `ultracode`.

### Request Example
```json
{
  "model": "anthropic/claude-opus-4.8",
  "max_tokens": 1024,
  "messages": [
    { "role": "user", "content": "Explain recursion in one sentence." }
  ],
  "stream": false
}
```

### Response
#### Success Response (200)
- **id** (string) - Unique identifier for the message.
- **type** (string) - Type of the response.
- **role** (string) - Role of the sender (e.g., "assistant").
- **content** (array) - Content of the message.
  - **type** (string) - Type of content (e.g., "text").
  - **text** (string) - The text content.
- **model** (string) - The model used for generation.
- **usage** (object) - Token usage information.
  - **input_tokens** (integer) - Number of input tokens.
  - **output_tokens** (integer) - Number of output tokens.
  - **cache_read_input_tokens** (integer) - Tokens read from cache.
  - **cache_creation_input_tokens** (integer) - Tokens used for cache creation.

#### Response Example
```json
{
  "id": "msg_...",
  "type": "message",
  "role": "assistant",
  "content": [
    { "type": "text", "text": "Recursion is when a function calls itself..." }
  ],
  "model": "anthropic/claude-opus-4.8",
  "usage": {
    "input_tokens": 14,
    "output_tokens": 32,
    "cache_read_input_tokens": 0,
    "cache_creation_input_tokens": 0
  }
}
```

### Streaming
Set `"stream": true` to receive server-sent events (SSE). The event format follows the Anthropic streaming spec.

### Extended Thinking
Supported via the `anthropic-beta: interleaved-thinking-2025-05-14` header. See Anthropic's extended thinking docs.
```
Tokens:
Raw