Skip to content

providers.md

Latest commit

 

History

History
26 lines (24 loc) · 2.77 KB

providers.md

File metadata and controls

26 lines (24 loc) · 2.77 KB
Raw

Currently available providers

  • AnyAPI (Multi-model API, 100k free anytokens/day, supports chat and image gen, many providers including deepseek, google, openai)
  • Atlas Cloud (OpenAI-compatible API, requires ATLASCLOUD_API_KEY, default model qwen/qwen3.8-max, supports ATLASCLOUD_MODEL, ATLASCLOUD_URL, and ATLASCLOUD_BASE_URL)
  • Deepseek (Requires API key)
  • Deepseek Web (Web API, requires user token from chat.deepseek.com via DEEPSEEK_WEB_TOKEN / --key and a JS runtime like node, bun, or deno in PATH to solve Proof-of-Work, supports optional DEEPSEEK_WEB_THINKING=true and DEEPSEEK_WEB_SEARCH=true env vars)
  • Fx (Free, fx.sh gateway, default model zai/glm-5.2)
  • Groq (Requires a free API Key. Many models)
  • Isou (Free) (Deepseek-chat with SEARXNG)
  • KoboldAI (Free) (koboldcpp/HF_SPACE_Tiefighter-13B)
  • LiteLLM (AI gateway proxy, access 100+ LLM providers via unified OpenAI-compatible API, default endpoint http://localhost:4000/v1/chat/completions, requires --model flag or LITELLM_MODEL env var)
  • MiniMax (Requires API key, default model MiniMax-M2.7)
  • Ollama (Local models) (Supports many models)
  • OllamaCloud (Cloud API) (Requires OLLAMA_API_KEY, default model gpt-oss:120b)
  • OmniRoute (OpenAI-compatible API router, default base URL http://localhost:20128/v1, default model auto, supports OMNIROUTE_API_KEY, OMNIROUTE_MODEL, OMNIROUTE_URL, OMNIROUTE_BASE_URL env vars)
  • OpenAI (All models, Requires API Key, supports custom endpoints)
  • OpenCode (Free, uses opencode.ai/zen API, default model deepseek-v4-flash-free, default API key public, supports OPENCODE_API_KEY / OPENCODE_MODEL / OPENCODE_URL env vars)
  • OpenRouter (OpenAI-compatible API, default model openrouter/free, supports OPENROUTER_API_KEY, OPENROUTER_MODEL, OPENROUTER_URL, and OPENROUTER_BASE_URL env vars)
  • Pollinations (Many free models)
  • Gemini (Require a free API keys, supports many models, default model gemini-2.0-flash)
  • PowerBrain (Free)

Image Generation Models:

  • AnyAPI (Requires API key, 100k anytokens/day)
  • Pollinations (Free) (Models)
  • MagicStudio (Free)