Repository navigation
Conversation
Zen no longer serves muse-spark-1.3-contributor-free, so any request without a model field returned 400 model_not_found. The default is now deepseek-v4.1-flash-free, which still handles tools and images. prepare_turn and list_models resolve the default against the live Zen catalogue, so a retired id gets swapped for another free model and logs a warning instead of breaking every caller. Explicit model ids are unchanged and unknown ones still return 400 with suggestions. Version 0.4.2.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Zen dropped muse-spark-1.3-contributor-free from the free list. Any request that didn't name a model got
400 model_not_found, which is what most simple clients send.Changes:
deepseek-v4.1-flash-free.prepare_turnandlist_modelscheck the default against the live Zen list. If it's gone, they swap in another free model and log a warning instead of failing the request.Explicit ids behave the same. A typo or a retired id sent explicitly still returns 400 with suggestions, since that's a caller mistake worth surfacing.
Tested with the release binary: default chat, streaming, image input, tool calls, a stale
WRAP_MODEL, and an explicit bad id. 44 unit tests pass, clippy and fmt clean.