feat: add claude-fable-5 and claude-fable-5-1 - #35
Merged
Merged
Conversation
Add the compiled-in constants `.fable5` and `.fable5_1`. Both take the same request surface as `.opus5`: every effort level (`low` through `max`), adaptive thinking, structured output and image input, with sampling parameters rejected. The capability matrix comment now groups them with Opus 5. `claude-fable-5-1` additionally rejects forced tool use (`tool_choice` of type `any` or `tool`). The bridge treats a `.required` tool-calling mode as a contract rather than a hint, so it is still sent as-is and the API error names the field; the comment says so.
dtmeadows-ant
approved these changes
Sep 9, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds the compiled-in constants
.fable5(claude-fable-5) and.fable5_1(claude-fable-5-1).Both take the same request surface as
.opus5: every effort level (lowthroughmax), adaptive thinking, structured output and image input, with sampling parameters rejected.claude-fable-5-1additionally rejects forced tool use (tool_choiceof typeanyortool); a.requiredtool-calling mode is a contract rather than a hint, so the bridge still sends it as-is and the API error names the field. The capability matrix comment says so.These are the models that server-side fallbacks (#32) exist for: they decline requests in certain policy areas, and
claude-opus-5andclaude-opus-4-8are their permitted fallback targets.Test plan
Live requests in the shapes these constants produce all return 200 on both models: adaptive thinking with summarized display, each of the five effort levels, a
json_schemaoutput format (the response conforms), and an image content block. Both models rejecttemperaturewith a 400, matchingsamplingParams: false.tool_choice: {"type": "any"}returns 200 onclaude-fable-5and 400 onclaude-fable-5-1.Through
LanguageModelSession, both constants answered at effortmax, honored a@Generableschema, and streamed reasoning entries.The full test suite at 7313964, via
xcodebuildon macOS 26.6 with Xcode 27.0 beta 6 (27A5252f) and the iOS 27.0 simulator:** TEST SUCCEEDED **, 243 tests, 240 passed, 0 failed, 3 skipped. The 3 skips are the keychain-backedAppAttestStoreTests, which skip by design under this runner.