​

Model Providers

​

クイックスタート

1. providerで認証します（通常は openclaw onboard 経由）。
2. デフォルトmodelを設定します:

{
  agents: { defaults: { model: { primary: "anthropic/claude-opus-4-6" } } },
}

​

Providerドキュメント

• Amazon Bedrock
• Anthropic (API + Claude CLI)
• BytePlus (International)
• Chutes
• Cloudflare AI Gateway
• DeepSeek
• Fireworks
• GitHub Copilot
• GLM models
• Google (Gemini)
• Groq (LPU inference)
• Hugging Face (Inference)
• Kilocode
• LiteLLM (unified gateway)
• MiniMax
• Mistral
• Moonshot AI (Kimi + Kimi Coding)
• NVIDIA
• Ollama (cloud + local models)
• OpenAI (API + Codex)
• OpenCode
• OpenCode Go
• OpenRouter
• Perplexity (web search)
• Qianfan
• Qwen Cloud
• Qwen / Model Studio（endpoint detail; qwen-* が正規、modelstudio-* はレガシー）
• SGLang (local models)
• StepFun
• Synthetic
• Together AI
• Venice (Venice AI, privacy-focused)
• Vercel AI Gateway
• vLLM (local models)
• Volcengine (Doubao)
• xAI
• Xiaomi
• Z.AI

​

共有概要ページ

• Additional bundled variants - Anthropic Vertex、Copilot Proxy、Gemini CLI OAuth

​

Transcription providers

• Deepgram (audio transcription)

​

Community tools

• Claude Max API Proxy - Claude subscription認証情報向けcommunity proxy（使用前にAnthropicのpolicy/termsを確認してください）