Gateway API

Use the local OpenAI-compatible and Anthropic-compatible endpoints from any app, SDK, or CLI tool.

CrabDash runs a CrabLLM gateway on 127.0.0.1:5635. It exposes both an OpenAI-compatible and an Anthropic-compatible API — use whichever format your tools expect.

Endpoints

Format	Base URL
OpenAI-compatible	`http://127.0.0.1:5635/v1`
Anthropic-compatible	`http://127.0.0.1:5635/anthropic`

No API key is required by default (the gateway is local-only). If you enable virtual keys in settings, pass them via the Authorization header as usual.

Supported routes

Route	Description
`GET /v1/models`	List available models across all configured providers
`POST /v1/chat/completions`	OpenAI-format chat completions (streaming and non-streaming)
`POST /anthropic/v1/messages`	Anthropic-format messages (streaming and non-streaming)
`GET /docs`	OpenAPI documentation for the full gateway API

For the complete API reference, open http://127.0.0.1:5635/docs in your browser while CrabDash is running, or see the CrabLLM API docs.

Usage with SDKs

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="http://127.0.0.1:5635/v1",
    api_key="unused",
)

response = client.chat.completions.create(
    model="claude-sonnet-4-20250514",
    messages=[{"role": "user", "content": "Hello"}],
)

Python (Anthropic SDK)

import anthropic

client = anthropic.Anthropic(
    base_url="http://127.0.0.1:5635/anthropic",
    api_key="unused",
)

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)

Node.js (OpenAI SDK)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "http://127.0.0.1:5635/v1",
  apiKey: "unused",
});

const response = await client.chat.completions.create({
  model: "gpt-4o",
  messages: [{ role: "user", content: "Hello" }],
});

Node.js (Anthropic SDK)

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

const client = new Anthropic({
  baseURL: "http://127.0.0.1:5635/anthropic",
  apiKey: "unused",
});

const message = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello" }],
});

curl (OpenAI format)

curl http://127.0.0.1:5635/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.0-flash",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'

curl (Anthropic format)

curl http://127.0.0.1:5635/anthropic/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: unused" \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Pointing apps at CrabDash

Any app that lets you set a custom base URL works with CrabDash. Use whichever format the app expects:

OpenAI-format apps: Set base URL to http://127.0.0.1:5635/v1
Anthropic-format apps: Set base URL to http://127.0.0.1:5635/anthropic

Common examples:

Continue (VS Code): Set apiBase to http://127.0.0.1:5635/v1 in your model config
Cursor: Settings → Models → OpenAI API Base → http://127.0.0.1:5635/v1
aider: aider --openai-api-base http://127.0.0.1:5635/v1
Claude Code: ANTHROPIC_BASE_URL=http://127.0.0.1:5635/anthropic claude

Model routing

When multiple providers serve the same model, CrabDash routes using weighted random selection with automatic failover. Configure weights and routing rules in Settings → Routing.

For full routing configuration details, see CrabLLM Configuration.

Gateway API

On this page