fix(models): include Muse Spark 1.2 variants in offline vision fallback (#183) - #189
Merged
ltmoerdani merged 1 commit intoAug 23, 2026
Conversation
…ck (ltmoerdani#183) models.dev lists image/video/pdf/audio input modalities for both muse-spark-1.2-contributor and -free, but VISION_CAPABLE_MODELS (the offline fallback) lacked them — so when neither live metadata nor the models.dev snapshot was available, the picker reported imageInput: false and VS Code stripped image attachments before they reached the provider. Note: the primary ltmoerdani#183 scenario (v0.6.0 sending muse through chat-completions) was already fixed by 5586836, which routes muse to the Responses endpoint in 0.7.0+.
Owner
|
nice one, verified locally and this checks out. What I confirmed on my end:
Small comment on the diff itself: the contract note you added above the two entries is helpful, no change requested there. Merging this now. Thanks for keeping the offline fallback path in mind, most people only test with live metadata available. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #183
PEACE BE UPON YOU
Summary
Addresses #183 (Muse Spark 1.2 Contributor Free on Zen + images). Two parts:
1. Already fixed in 0.7.0 — please upgrade. The reporter's stack trace is from v0.6.0, where
muse-spark-1.2-contributor-freewent through the chat-completions endpoint — which the Muse gateway rejects (that's why the 400 body is a chat.completion object with an empty message andfinish_reason: null). Commit 5586836 (released in 0.7.0) routes allmuse-*models to the Responses API, and images are mapped to the correctinput_imageshape there.2. This PR — offline vision fallback gap. models.dev lists image/video/pdf/audio input modalities for both Muse variants, but
VISION_CAPABLE_MODELS(the bundled fallback used when neither live metadata nor a cached models.dev snapshot is available) lacked them. In that state the picker reportedimageInput: false, so VS Code stripped image attachments before they ever reached the provider. Both variants are now in the set, so offline/cold-start users can attach images to Muse models too.Verification
npm run lintall 7 checks green; new assertion covers both variants.Fi Amanillah