Summary
PR #12046 added support for dynamically fetching provider models. This is a significant new feature that allows users to:
- Auto-load models for
ollama and openrouter providers
- See a refresh button to fetch provider models after entering an API key
- Have fetched models merged and deduplicated with hardcoded models
- Persist
contextLength, completionOptions.maxTokens, and capabilities to config.yaml when adding a model
Documentation Needed
- Document how the dynamic model fetching works
- Explain the refresh button functionality in the Add New Model UI
- Document which providers support dynamic fetching (
ollama, openrouter, anthropic, gemini, and others via listModels())
- Explain how fetched model properties are persisted to config
Related PR
Generated with Continue
Co-Authored-By: Continue noreply@continue.dev
Co-authored-by: bekahhw rhawrot@gmail.com
Summary
PR #12046 added support for dynamically fetching provider models. This is a significant new feature that allows users to:
ollamaandopenrouterproviderscontextLength,completionOptions.maxTokens, andcapabilitiestoconfig.yamlwhen adding a modelDocumentation Needed
ollama,openrouter,anthropic,gemini, and others vialistModels())Related PR
Generated with Continue
Co-Authored-By: Continue noreply@continue.dev
Co-authored-by: bekahhw rhawrot@gmail.com