Repository navigation
v3.11.0: Claude models from the Models API, benchmark log buttons, per-drawing chat keeps choices - #24
Merged
Conversation
…nchmark log buttons - ClaudeModelCatalog reads GET /v1/models (cached 5 min, Refresh forces a re-read); without an API key the current generation (Fable 5.2, Opus 5.5, Sonnet 5.5, Haiku 5.5) is shown instead of Sonnet 5 / Opus 4.8. - API and Claude Code pickers and the benchmark window use the catalog; the Refresh models button now works for Claude agents too. - Auto defaults move to Sonnet 5.5 executor + Opus 5.5 advisor (Opus 4.8 is rejected as advisor by Sonnet 5.5); older saved choices migrate to the same tier. - Model rules key on family/version, so any served id gets pricing, thinking and history rules; Fable 5.2 added. - Benchmark window: Open log and Log folder buttons for the per-run JSONL log. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01C8csiSMaAtaRVzrSAs372U
A drawing's first chat session now inherits the current agent, model and effort choices instead of falling back to Auto, and an unsent draft moves with the user. Chats stay per drawing; attachments stay with their drawing. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01C8csiSMaAtaRVzrSAs372U
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01C8csiSMaAtaRVzrSAs372U
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Claude models are loaded like the Codex catalog
ClaudeModelCatalogreadsGET /v1/modelsfor the API key (5-minute cache; Refresh models forces a re-read and now works for Claude agents too). Without a key or on error, the current generation is shown: Fable 5.2, Opus 5.5, Sonnet 5.5, Haiku 5.5 — no more Sonnet 5 / Opus 4.8 defaults.opus/sonnetaliases.ClaudeModelsrules (pricing, adaptive thinking, append-only history, server fallback) are keyed by family/version, so any served id is handled; Fable 5.2 added.Benchmark window
Chat per drawing
Release notes:
docs/releases/v3.11.0.md.Testing
ClaudeModelsTests) and pane checks updated for the catalog, id parsing, pricing and new picker values.Generated by Claude Code