Add Claude Sonnet 5.5 model support - #122
Merged
Merged
Conversation
- Capability rule for claude-sonnet-5-5 (128K output, 1M context), placed before the generic claude-sonnet-5 rule. Adaptive thinking is on by default and `thinking: disabled` returns a 400, so omitting thinking cannot turn it off; "off" is not offered (Opus 5.5 precedent). - Supplemental model entry so Sonnet 5.5 appears in the Anthropic selector until /v1/models returns it. - No Posit AI Pass supersession: Sonnet 5 stays selectable (it moves to the picker's More models submenu) since 5.5 has breaking API changes.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds Claude Sonnet 5.5 (
claude-sonnet-5-5, released 2026-09-28) to the model catalog, so it shows up with the right context limits and thinking-effort options on Anthropic, Posit AI Pass, Bedrock, and OpenRouter-style IDs.ai-config): 1M context, 128K max output, familyclaude-5.5. It sits before the broader^claude-sonnet-5rule, which would otherwise match it.thinking: {type: "disabled"}returns a 400. Our "off" works by leaving thethinkingfield out, which on this model would still think, at the API's default effort, with thinking text omitted. So "off" isn't offered, same as for Opus 5.5. The model's real lowest setting,thinking: {type: "between_tools"}, isn't wired into the clients./v1/modelsreturns it.Companion monorepo PR: pins the submodule to this change and adds pricing and model-picker defaults.
Verification
claude-sonnet-5still gets the Sonnet 5 rule with "off".ai-configmodel-capability tests,ai-provider-bridgeprovider tests, and type checks pass.