OpenClaw can use many LLM providers. Pick one, authenticate, then set the default
model as provider/model.
Quick start (two steps)
- Authenticate with the provider (usually via
openclaw onboard). - Set the default model:
{
agents: { defaults: { model: { primary: "anthropic/claude-opus-4-6" } } },
}
Supported providers (starter set)
- [Alibaba Model Studio](/docs/openclaw-docs/providers/alibaba
- [Amazon Bedrock](/docs/openclaw-docs/providers/bedrock
- [Anthropic (API + Claude CLI)](/docs/openclaw-docs/providers/anthropic
- [BytePlus (International)](/docs/openclaw-docs/concepts/model-providers#byteplus-international
- [Chutes](/docs/openclaw-docs/providers/chutes
- [ComfyUI](/docs/openclaw-docs/providers/comfy
- [Cloudflare AI Gateway](/docs/openclaw-docs/providers/cloudflare-ai-gateway
- [DeepInfra](/docs/openclaw-docs/providers/deepinfra
- [fal](/docs/openclaw-docs/providers/fal
- [Fireworks](/docs/openclaw-docs/providers/fireworks
- [GLM models](/docs/openclaw-docs/providers/glm
- [MiniMax](/docs/openclaw-docs/providers/minimax
- [Mistral](/docs/openclaw-docs/providers/mistral
- [Moonshot AI (Kimi + Kimi Coding)](/docs/openclaw-docs/providers/moonshot
- [OpenAI (API + Codex)](/docs/openclaw-docs/providers/openai
- [OpenCode (Zen + Go)](/docs/openclaw-docs/providers/opencode
- [OpenRouter](/docs/openclaw-docs/providers/openrouter
- [Qianfan](/docs/openclaw-docs/providers/qianfan
- [Qwen](/docs/openclaw-docs/providers/qwen
- [Runway](/docs/openclaw-docs/providers/runway
- [StepFun](/docs/openclaw-docs/providers/stepfun
- [Synthetic](/docs/openclaw-docs/providers/synthetic
- [Vercel AI Gateway](/docs/openclaw-docs/providers/vercel-ai-gateway
- [Venice (Venice AI)](/docs/openclaw-docs/providers/venice
- [xAI](/docs/openclaw-docs/providers/xai
- [Z.AI](/docs/openclaw-docs/providers/zai
Additional bundled provider variants
anthropic-vertex- implicit Anthropic on Google Vertex support when Vertex credentials are available; no separate onboarding auth choicecopilot-proxy- local VS Code Copilot Proxy bridge; useopenclaw onboard --auth-choice copilot-proxygoogle-gemini-cli- unofficial Gemini CLI OAuth flow; requires a localgeminiinstall (brew install gemini-cliornpm install -g @google/gemini-cli); default modelgoogle-gemini-cli/gemini-3-flash-preview; useopenclaw onboard --auth-choice google-gemini-clioropenclaw models auth login --provider google-gemini-cli --set-default
For the full provider catalog (xAI, Groq, Mistral, etc.) and advanced configuration, see [Model providers](/docs/openclaw-docs/concepts/model-providers.
Related
- [Model selection](/docs/openclaw-docs/concepts/model-providers
- [Model failover](/docs/openclaw-docs/concepts/model-failover
- [Models CLI](/docs/openclaw-docs/cli/models