Model selector
The model selector at the top of the page shows the currently active model. Click to switch between any configured model — built-in or custom. Each agent instance can use a different model.Built-in providers
For each built-in provider, choose how to connect:| Provider | Models |
|---|---|
| Anthropic | Claude Opus, Sonnet, Haiku |
| OpenAI | GPT and Codex models |
| Gemini models | |
| Moonshot AI | Kimi models |
| Alibaba Cloud | Qwen models |
| DeepSeek | DeepSeek V-series models |
| Z.AI | GLM models |
| MiniMax | MiniMax M-series models |
stagewise Account
Use the managed inference service with a curated model library. No API keys needed. Works out of the box.Import your subscription (BYOK)
Already paying for a provider? Enter your API key to route requests directly to their API. Set up BYOK →Custom endpoint
Connect to any compatible service — including models running locally on your machine. Connect a custom provider →Custom models
Below the built-in providers, the Custom Models section lets you add models not in the default list. Click Add model and provide:- Model ID — The identifier your endpoint expects
- Display name — How it appears in the model selector
- Provider — Which built-in provider this model routes through