Repository navigation
fix(mistral): treat model_length as a context window finish reason - #477
Open
pcbeingused333 wants to merge 3 commits into
Open
pcbeingused333 wants to merge 3 commits into
pcbeingused333 wants to merge 3 commits into
Conversation
Mistral stops a response with finish_reason model_length when the request reaches the model's context length. It was passed through as an unknown raw string, so chatlas neither warned about the truncated response nor refused to parse truncated structured data. Map it to "context_window". Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
# Conflicts: # CHANGELOG.md # tests/test_provider_openai_completions.py
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Mistral ends a response with
finish_reason: "model_length"when the request reaches the model's context length.normalize_finish_reasonin the Chat Completions provider, whichChatMistral()uses, did not know the value and passed the raw string through. chatlas therefore did not warn that the response was truncated, and.chat_structured()tried to parse the cut-off JSON instead of failing with the truncation message, which is what happens for the other providers' context-window reasons.model_lengthnow maps to"context_window", the value Anthropic and Bedrock already use for the same situation.Tests.
test_normalize_finish_reason_maps_known_reasonscovers the mapping, andtest_model_length_finish_reason_warns_about_the_context_windowchecks that amodel_lengthcompletion becomes acontext_windowturn and raises the context-window error when a data model was requested. Both fail on main and pass here, and the rest oftests/test_provider_openai_completions.pyandtests/test_provider_mistral.pypass. No request was made to the Mistral API.This change was made with Claude Code.
🤖 Generated with Claude Code