TriCode Studio blogGuides and notes from TriCode Studio.https://tricodestudio.com/Local LLMs for meeting notes: we tested 0.8B to 9B on real meetingshttps://tricodestudio.com/blog/local-llm-meeting-notes-benchmark/https://tricodestudio.com/blog/local-llm-meeting-notes-benchmark/We ran Qwen3.5 0.8B–9B, Gemma 4 E2B and LFM2.5 on real AMI meetings on a 16 GB Mac, and measured memory, speed and how many human-noted decisions each one caught.Thu, 08 Oct 2026 00:00:00 GMTlocal llmllama.cppqwenmeeting notesbenchmark