Summary
PR #12046 merged a feature that allows users to dynamically fetch provider models using the models/fetch protocol. This feature:
- Auto-loads models for
ollama and openrouter providers
- Adds a refresh button that appears after entering an API key to fetch available models from providers
- Supports fetching models from
anthropic, gemini, and other providers using their listModels methods
- Persists
contextLength, maxTokens, and capabilities to config.yaml when adding a model
Documentation Needed
The documentation should cover:
- How to use the refresh/fetch models feature in the Add Model UI
- Which providers support dynamic model fetching
- What model properties are automatically detected and persisted
- Any limitations or considerations when using this feature
Suggested Location
docs/customize/models.mdx - Add a section about discovering available models
- Or create a new guide under
docs/guides/
Related PR
Generated with Continue
Co-Authored-By: Continue noreply@continue.dev
Co-authored-by: bekahhw rhawrot@gmail.com
Summary
PR #12046 merged a feature that allows users to dynamically fetch provider models using the
models/fetchprotocol. This feature:ollamaandopenrouterprovidersanthropic,gemini, and other providers using theirlistModelsmethodscontextLength,maxTokens, andcapabilitiestoconfig.yamlwhen adding a modelDocumentation Needed
The documentation should cover:
Suggested Location
docs/customize/models.mdx- Add a section about discovering available modelsdocs/guides/Related PR
Generated with Continue
Co-Authored-By: Continue noreply@continue.dev
Co-authored-by: bekahhw rhawrot@gmail.com