Document dynamic provider model fetching feature

[continue]

Summary

PR #12046 introduced a feature for dynamically fetching provider models but there is no documentation for it.

Missing Documentation

The following features need to be documented:

• The new refresh icon that appears after entering an API key to "update models list using my API key"
• Support for dynamically fetching models from various providers:
  • Ollama (using their tags API)
  • OpenRouter
  • Anthropic (/v1/models)
  • Gemini (v1beta/models)
  • Other providers using listModels() methods
• How the system persists fetched contextLength, maxTokens, and capabilities to config.yaml when adding a model
• New model support: GPT-5.4 (pro/mini) and Gemma 4

Related PR

• feat: fetch provider models dynamically and fix isOpenSource pollution #12046

Suggested Location

This documentation could be added to:

• docs/customize/models.mdx
• Provider-specific pages in docs/customize/model-providers/

---

Generated with Continue

Co-Authored-By: Continue noreply@continue.dev
Co-authored-by: bekahhw rhawrot@gmail.com