fix: fallback to Ollama native API for thinking-mode models#27
Open
wbryanta wants to merge 1 commit intonikmcfly:mainfrom
Open
fix: fallback to Ollama native API for thinking-mode models#27wbryanta wants to merge 1 commit intonikmcfly:mainfrom
wbryanta wants to merge 1 commit intonikmcfly:mainfrom
Conversation
Models with thinking mode (e.g., Gemma 4) return empty content via Ollama's OpenAI-compatible /v1/chat/completions endpoint. The thinking tokens consume the max_tokens budget but visible content is stripped. This adds a fallback in LLMClient.chat(): when the OpenAI endpoint returns empty content and we're talking to Ollama, retry via the native /api/chat endpoint which handles thinking tokens correctly. Backwards-compatible — only triggers on empty responses. Fixes nikmcfly#26 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #26
Problem
Models with thinking mode (e.g., Gemma 4) return empty
contentvia Ollama's/v1/chat/completionsendpoint. Thinking tokens consume themax_tokensbudget but visible content is stripped, causing 500 errors on simulation start.Fix
When
LLMClient.chat()gets empty content from the OpenAI-compatible endpoint and the backend is Ollama, it retries via the native/api/chatendpoint, which handles thinking tokens correctly.requestsalready in requirements)gemma4:26bon Ollama 0.20.0