Release v0.12.8 · ollama/ollama

What's Changed qwen3-vl performance improvements, including flash attention support by default qwen3-vl will now output less leading whitespace in the response when thinking Fixed issue where ...

Ollama 0.12.8

What's Changed

qwen3-vl performance improvements, including flash attention support by default

qwen3-vl will now output less leading whitespace in the response when thinking

Fixed issue where deepseek-v3.1 thinking could not be disabled in Ollama's new app

Fixed issue where qwen3-vl would fail to interpret images with transparent backgrounds

Ollama will now stop running a model before removing it via ollama rm

Fixed issue where prompt processing would be slower on Ollama's engine

Ignore unsupported iGPUs when doing device discovery on Windows

Full Changelog: 

https://github.com/ollama/ollama/releases/tag/v0.12.8

Ollama · GitHub  https://github.com/ollama