ggml/llama.cpp releases: b12019

TL;DR

The ggml/llama.cpp project updated its codebase to include changes in how the Qwen3-coder handles think blocks, either stripping them or accepting them based on the reasoning_format setting. This update is significant for developers who need to customize the response format of the Qwen3-coder AI model.

Detailed Summary

The ggml/llama.cpp project updated its codebase to include changes in chat responses for the qwen3-coder, specifically stripping or accepting a "think" block with reasoning_format none. These updates were assisted by Claude Opus 5.5 and involve defining both variants of the think block upfront, impacting how the qwen3-coder processes and formats its responses.

Key Points

• • chat : strip the think block in the qwen3-coder with reasoning_format none
• • chat : accept the think block in the qwen3-coder response format with reasoning_format none
• • chat : define both qwen3-coder think block variants up front