Integrations

Works with what you're already using.
Setup is a base URL.

OpenAI-compatible at the endpoint means every SDK, framework, and coding agent in the ecosystem routes through us without a rewrite. Copy the snippet, change one string, ship.

Read the docsFor coding agents
Routes from 29+ surfaces
OpenAI SDK
Claude Code
Cursor
Cline
Aider
Continue.dev
LangChain
LlamaIndex
Vercel AI SDK
LiteLLM
MCP (Model Context Protocol)
Vercel
Anthropic SDK
Mastra
OpenClaw
Hermes
n8n
AIgateway Python SDK
AIgateway Node SDK
AIgateway CLI
Promptfoo
LobeChat
LibreChat
Open WebUI
Zed
PA
Pydantic AI
CR
CrewAI
AG
AutoGen
LF
Langfuse
Featured

Start here.

The three surfaces most teams plug in first. Every other integration follows the same pattern: base URL + your AIgateway key.

Claude Code
CLI agent
Point Anthropic's CLI at AIgateway and pick any model in 30 seconds.
export ANTHROPIC_BASE_URL="https://api.aigateway.sh"
export ANTHROPIC_API_KEY="sk-aig-..."
claude
Full setup→
Cursor
IDE
Add AIgateway as a custom OpenAI-compatible provider in Cursor.
# Cursor → Settings → Models → Custom
Base URL:  https://api.aigateway.sh/v1
API key:   sk-aig-...
Models:    anthropic/claude-opus-4.7,
           openai/gpt-5.4,
           moonshot/kimi-k2.6
Full setup→
OpenAI SDK
SDK
Drop-in. Change one line, gain 1000+ models.
from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)
print(client.chat.completions.create(
    model="anthropic/claude-opus-4.7",
    messages=[{"role":"user","content":"hi"}],
))
Full setup→
All integrations

Filter, search, pick your stack.

Official SDKs, drop-in clients, coding agents, chat UIs, orchestration frameworks. 29 surfaces, all OpenAI-compatible at the wire.

OpenAI SDK
SDK
Drop-in. Change one line, gain 1000+ models.
from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)
print(client.chat.completions.create(
    model="anthropic/claude-opus-4.7",
    messages=[{"role":"user","content":"hi"}],
))
Full setup→
Claude Code
CLI agent
Point Anthropic's CLI at AIgateway and pick any model in 30 seconds.
export ANTHROPIC_BASE_URL="https://api.aigateway.sh"
export ANTHROPIC_API_KEY="sk-aig-..."
claude
Full setup→
Cursor
IDE
Add AIgateway as a custom OpenAI-compatible provider in Cursor.
# Cursor → Settings → Models → Custom
Base URL:  https://api.aigateway.sh/v1
API key:   sk-aig-...
Models:    anthropic/claude-opus-4.7,
           openai/gpt-5.4,
           moonshot/kimi-k2.6
Full setup→
Cline
VS Code ext
VS Code agent that runs on whichever model wins your eval.
Provider:  OpenAI Compatible
Base URL:  https://api.aigateway.sh/v1
API key:   sk-aig-...
Model ID:  anthropic/claude-opus-4.7
Full setup→
Aider
CLI agent
Terminal-native pair programmer — works with one --openai-api-base flag.
export OPENAI_API_BASE="https://api.aigateway.sh/v1"
export OPENAI_API_KEY="sk-aig-..."
aider --model anthropic/claude-opus-4.7
Full setup→
Continue.dev
IDE ext
Open-source IDE assistant — drop AIgateway into config.json.
{
  "models": [{
    "title": "AIgateway",
    "provider": "openai",
    "apiBase": "https://api.aigateway.sh/v1",
    "apiKey": "sk-aig-...",
    "model": "anthropic/claude-opus-4.7"
  }]
}
Full setup→
LangChain
Library
ChatOpenAI + AIgateway base URL = every model, zero new code.
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="anthropic/claude-opus-4.7",
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)
print(llm.invoke("hi").content)
Full setup→
LlamaIndex
Library
RAG framework — point its OpenAI provider at AIgateway.
from llama_index.llms.openai import OpenAI
from llama_index.core import Settings

Settings.llm = OpenAI(
    api_base="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
    model="openai/gpt-5.4",
)
Full setup→
Vercel AI SDK
Library
Use @ai-sdk/openai with a custom baseURL — every model, every modality.
import { createOpenAI } from "@ai-sdk/openai";
import { streamText } from "ai";

const ai = createOpenAI({
  baseURL: "https://api.aigateway.sh/v1",
  apiKey: process.env.AIGATEWAY_API_KEY,
});

const r = await streamText({
  model: ai("anthropic/claude-opus-4.7"),
  prompt: "hi",
});
Full setup→
LiteLLM
Proxy
Use LiteLLM as a proxy upstream of AIgateway.
model_list:
  - model_name: kimi-k2.6
    litellm_params:
      model: openai/moonshot/kimi-k2.6
      api_base: https://api.aigateway.sh/v1
      api_key: os.environ/AIGATEWAY_API_KEY
Full setup→
MCP (Model Context Protocol)
MCP server
Mount AIgateway as an MCP server inside any compatible host.
{
  "mcpServers": {
    "aigateway": {
      "type": "http",
      "url": "https://api.aigateway.sh/mcp",
      "headers": { "Authorization": "Bearer sk-aig-..." }
    }
  }
}
Full setup→
Vercel
Deploy
One env var → every Vercel AI app uses AIgateway.
AIGATEWAY_API_KEY=sk-aig-...
OPENAI_BASE_URL=https://api.aigateway.sh/v1
Full setup→
Anthropic SDK
SDK
Official Anthropic client. Swap base URL, keep everything else.
import anthropic

client = anthropic.Anthropic(
    base_url="https://api.aigateway.sh/anthropic",
    api_key="sk-aig-...",
)
msg = client.messages.create(
    model="claude-opus-4.7",
    max_tokens=1024,
    messages=[{"role":"user","content":"hi"}],
)
Full setup→
Mastra
Framework
Agent framework — point models at the gateway, tools route automatically.
import { Agent } from "@mastra/core/agent";
import { openai } from "@ai-sdk/openai";

const agent = new Agent({
  name: "researcher",
  model: openai("anthropic/claude-opus-4.7", {
    baseURL: "https://api.aigateway.sh/v1",
  }),
});
Full setup→
OpenClaw
CLI agent
Autoconfigure from llms.txt — one command.
openclaw config add aigateway \
  --llms-txt https://aigateway.sh/llms.txt \
  --api-key sk-aig-...
Full setup→
Hermes
Autonomous agent
Nous Research's self-hosted agent — one config block points it at any model.
# ~/.hermes/config.yaml
model:
  default: moonshot/kimi-k2.6
  provider: custom
  base_url: https://api.aigateway.sh/v1
  api_key: sk-aig-...
Full setup→
n8n
Workflow
OpenAI credential → any model across every node.
n8n → Credentials → OpenAI API
Base URL:  https://api.aigateway.sh/v1
API key:   sk-aig-...
Full setup→
AIgateway Python SDK
Python SDK
Typed Python SDK for the aggregator-native surface — async jobs, sub-accounts, evals, replays, webhooks.
pip install aigateway-py

from aigateway import AIgateway
client = AIgateway(api_key="sk-aig-...")

job = client.jobs.create_video(
    model="runwayml/gen-4",
    prompt="a sunset over mountains",
)
done = client.jobs.wait(job.id)
Full setup→
AIgateway Node SDK
Node SDK
Typed TypeScript SDK for async jobs, sub-accounts, evals, replays, and webhook verification.
pnpm add aigateway-js

import { AIgateway } from "aigateway-js";
const ai = new AIgateway({ apiKey: process.env.AIGATEWAY_API_KEY! });

const acct = await ai.subAccounts.create({
  name: "acme-corp",
  spendCapCents: 50_000,
});
Full setup→
AIgateway CLI
CLI
`aig` — call any model, mint keys, run evals, tail traffic from your terminal.
npm i -g aigateway-cli

aig init
aig call moonshot/kimi-k2.6 "explain edge inference"
aig models --modality text
aig tail --filter status=error
Full setup→
Promptfoo
Eval
Eval prompts across every AIgateway model with one provider config.
providers:
  - id: openai:chat:anthropic/claude-opus-4.7
    config: { apiBaseUrl: "https://api.aigateway.sh/v1" }
  - id: openai:chat:moonshot/kimi-k2.6
    config: { apiBaseUrl: "https://api.aigateway.sh/v1" }
Full setup→
LobeChat
Chat UI
Self-hosted ChatGPT clone — one env var and every AIgateway model shows up.
docker run -d -p 3210:3210 \
  -e OPENAI_PROXY_URL=https://api.aigateway.sh/v1 \
  -e OPENAI_API_KEY=sk-aig-... \
  lobehub/lobe-chat
Full setup→
LibreChat
Chat UI
YAML config → every AIgateway model available as a LibreChat endpoint.
endpoints:
  custom:
    - name: "AIgateway"
      apiKey: ${AIGATEWAY_API_KEY}
      baseURL: https://api.aigateway.sh/v1
      models: { fetch: true }
Full setup→
Open WebUI
Chat UI
The 100k-star local ChatGPT — two env vars and you're on every AIgateway model.
docker run -d -p 3000:8080 \
  -e OPENAI_API_BASE_URL=https://api.aigateway.sh/v1 \
  -e OPENAI_API_KEY=sk-aig-... \
  ghcr.io/open-webui/open-webui:main
Full setup→
Zed
Editor
Native editor with AI baked in — drop AIgateway into settings.json.
{
  "language_models": {
    "openai_compatible": {
      "AIgateway": {
        "api_url": "https://api.aigateway.sh/v1",
        "available_models": [
          { "name": "anthropic/claude-opus-4.7", "max_tokens": 200000 }
        ]
      }
    }
  }
}
Full setup→
PA
Pydantic AI
Framework
Production-grade agent framework from the Pydantic team — wires up to AIgateway in one line.
from pydantic_ai import Agent
from pydantic_ai.models.openai import OpenAIModel
from pydantic_ai.providers.openai import OpenAIProvider

agent = Agent(OpenAIModel(
    "anthropic/claude-opus-4.7",
    provider=OpenAIProvider(
        base_url="https://api.aigateway.sh/v1",
        api_key="sk-aig-...",
    ),
))
Full setup→
CR
CrewAI
Framework
Multi-agent orchestration — route every Crew agent through AIgateway with a single env var.
export OPENAI_API_KEY="sk-aig-..."
export OPENAI_API_BASE="https://api.aigateway.sh/v1"
# CrewAI agents now use AIgateway across the catalog.
Full setup→
AG
AutoGen
Framework
Microsoft's multi-agent conversation framework — point any agent at AIgateway.
from autogen_ext.models.openai import OpenAIChatCompletionClient
client = OpenAIChatCompletionClient(
    model="anthropic/claude-opus-4.7",
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)
Full setup→
LF
Langfuse
Observability
Open-source LLM observability — trace every AIgateway call into Langfuse with one wrapper.
from langfuse.openai import OpenAI
client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)
# Every call below is traced to Langfuse.
Full setup→