event

LLM Fine-Tuning Announcement

4 items · started 2026-10-18 · closed 2026-10-25

  1. Oct 23, 2026 · 9 min read ThinkingCap: Gemma-4-26B-A4B — up to 69% and on average 39% less thinking, at unchanged accuracy TL;DR This is the fifth model in the ThinkingCap series. We reduced how much Google's gemma-4-26B-A4B-it thinks, wit…

  2. We study the 305M-parameter audio encoder shared by Gemma 4 and EmbeddingGemma 2 (identical except for its final projection) for Saudi and Gulf Arabic speech recognition, without changing any of its weights. A frozen system trains only a 2…

  3. After checking out Turbo Fieldfare (a very interesting project that streamed experts from SSD on Apple Silicon Macs to run Gemma 4 26B with very little RAM usage), I tried to make this project much faster and succeeded! This project gets u…

  4. I’m a machine learning engineer competing in the Gemma 4 fine-tuning challenge on Kaggle. I’m building a small synthetic software engineering dataset with a training/holdout split to fine-tune a small model.

← all threads