Models and providers
受支持供应商与模型目录
查看 OpenAI、Anthropic、Google Gemini、DeepSeek、xAI Grok、Ollama 和 llama.cpp 的集成信息。实际可用的具体模型取决于当前已配置的供应商及其同步结果。
Last reviewed:
Implemented cloud-provider adapters
The connector service includes operational adapters for OpenAI, Anthropic, Google Gemini, DeepSeek, and xAI Grok. Each adapter synchronizes the provider catalog when valid credentials and configuration are present. Model names and availability can change upstream, so the live catalog is the reliable source for a running deployment.
Local execution through Ollama and llama.cpp
Operators can configure Ollama and llama.cpp for models that run on controlled infrastructure. A local-only routing mode keeps its fallback chain on local providers. Hardware fit, model licensing, context limits, quality, and throughput vary, so private-deployment model selection is handled during technical scoping rather than promised globally.
Manual selection and policy-aware routing
People can choose an available model directly or allow routing logic to select an eligible path for the task. The router considers model capabilities, lifecycle, connector configuration, and policy. Multi-model modes add comparison and review patterns, but their availability and limits follow the active deployment rather than marketing copy.
Why this page avoids a frozen version list
A static list of fashionable model versions becomes inaccurate quickly and can imply credentials that are not configured. ClawAI therefore distinguishes implemented provider families from live model availability. Models shown inside the product are synchronized from enabled connectors and filtered by lifecycle and any administrator restrictions.
您获得的模型
一个工作区涵盖以下所有提供商。在对话中切换模型，或者让 ClawAI 为每条消息选择最好的模型。
OpenAI
当前可用 40 个模型
- GPT 6.1 Sol
- GPT 6.1 Terra
- GPT 6 Astra
- GPT 6 Luna
- GPT 6 Sol
- GPT 5.6 Luna
Anthropic
当前可用 15 个模型
- Claude Haiku 5.5
- Claude Opus 5.5
- Claude Sonnet 5.5
- Claude Fable 5.2
- Claude Fable 5.1
- Claude Fable 5
Google Gemini
当前可用 25 个模型
- Gemini 4 Argon
- Gemini 4 Carbon
- Gemini 3.8 Flash
- Gemini 3.7 Flash
- Gemini 3.6 Flash
- Gemini 3.5 Flash
Grok (xAI)
当前可用 12 个模型
- Grok 4.7
- Grok 4.6
- Grok 4.5
- Grok 4.3
- Grok 4.20 0309 Non Reasoning
- Grok 4.20 0309 Reasoning
Ollama
当前可用 23 个模型
- GLM 5.3
- GLM 5.3 Flash
- GLM 5.2
- Mistral Large 4
- Nemotron 3 Super
- Nemotron 3 Ultra
AI Horde
当前可用 2 个模型
- Gemma 4 E4B Uncensored HauhauCS Aggressive Q4 K M
- Qwen Qwen3 0.6B IQ4 XS
Groq
当前可用 4 个模型
- GPT OSS 120B
- Qwen3.8 27B
- GPT OSS 20B
- ALLaM 2 7b
Hugging Face
当前可用 80 个模型
- Kimi K2 Instruct 0905
- DeepSeek V4 Pro 0813
- DeepSeek V4 Flash 0731
- DeepSeek R1 0528
- Qwen3.5 397B A17B
- Qwen3 235B A22B Instruct 2507
Nvidia NIM
当前可用 66 个模型
- Palmyra Creative 122b
- Palmyra Fin 70b 32k
- Palmyra Med 70b 32k
- Codellama 70b
- llama2 70b
- Palmyra Med 70b
OpenRouter
当前可用 337 个模型
- Mistral: Mistral Small 4
- Mistral: Codestral 2508
- Mistral Large 2407
- MoonshotAI: Kimi K2 0905
- Qwen: Qwen3.8 Max (0902)
- DeepSeek: DeepSeek V4 Pro 0813
新的前沿模型在推出时就会被添加——您的计划从第一天起就涵盖它们，并根据单一限额进行计量。
Models are synchronized from enabled providers. The exact catalog depends on configured credentials, provider availability, lifecycle status, and administrator policy.
查看 OpenAI、Anthropic、Google Gemini、DeepSeek、xAI Grok、Ollama 和 llama.cpp 的集成信息。实际可用的具体模型取决于当前已配置的供应商及其同步结果。