Model Providers
WednesdayAI can use many LLM providers. Pick a provider, authenticate, then set the default model asprovider/model.
Looking for chat channel docs (WhatsApp/Telegram/Discord/Slack/Mattermost (plugin)/etc.)? See Channels.
Quick start
- Authenticate with the provider (usually via
wednesdayai onboard). - Set the default model:
Provider docs
- Amazon Bedrock
- Anthropic (API + Claude Code CLI)
- Cloudflare AI Gateway
- GLM models
- Hugging Face (Inference)
- Kilocode
- LiteLLM (unified gateway)
- MiniMax
- Mistral
- Moonshot AI (Kimi + Kimi Coding)
- NVIDIA
- Ollama (local models)
- OpenAI (API + Codex)
- OpenCode Zen
- OpenRouter
- Qianfan
- Qwen (OAuth)
- Together AI
- Vercel AI Gateway
- Venice (Venice AI, privacy-focused)
- vLLM (local models)
- Xiaomi
- Z.AI
Transcription providers
Audio/image/video understanding providers ship as plugin extensions, disabled by default. See Media Understanding for the full vendor list (OpenAI, Groq, Deepgram, Google, Mistral, ElevenLabs, OpenRouter, DeepInfra, Bifrost, and more) and how to enable one.Streaming voice conversation providers
The live, cascaded voice conversation loop behind the/v2/client web client’s Dictate and Voice
mode controls uses its own streaming speech-to-text/text-to-speech provider seam, separate from
the transcription providers above. See
Voice conversation sessions.
Community tools
- Claude Max API Proxy - Community proxy for Claude subscription credentials (verify Anthropic policy/terms before use)