» Tag
gemma
8 postsPorting Gemma-4 (2B/4B/12B) to AWS Inferentia2: A Field Report
How Gemma-4's mixed attention heads and KV-sharing broke AWS's vendor stack, and the tracing/sharding fix that made Inferentia2 inference work.
TurboFieldfare runs Gemma 4 26B MoE model in 2GB RAM on any Mac
TurboFieldfare is an open-source Swift/Metal runtime that streams MoE experts to run Gemma 4 26B in just 2GB RAM on 8GB Apple Silicon Macs.
Porting Gemma-4 (2B/4B/12B) to AWS Inferentia2: A Field Report
How Gemma-4 was ported to AWS Inferentia2 by bypassing vLLM/NxD, solving mixed attention heads and KV-sharing via direct model tracing.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comDiffusionGemma: Google's Diffusion-Based LLM Developer Guide
Google's DiffusionGemma developer guide: diffusion-based LLM with bidirectional attention, 4x faster generation, vLLM support, and Sudoku fine-tuning results.
A Cliché-Resistant Gemma Model for Storytelling
Gemma-4-26B-A4B-StyleTune-V2 is a language model designed for cliché-free storytelling.
Running Gemma 4 26B on a 13-Year-Old Xeon Without a GPU
Explore how to run Gemma 4 26B on an old Xeon server without a GPU.
Cactus Hybrid: Training Gemma 4 to Recognize Its Mistakes
Cactus has improved the Gemma 4 model to recognize its mistakes. The on-device model now provides reliable responses with confidence scores.
Trendyol's agent tunes LLM serving configs for 4x speedup
Trendyol Tech built autooptimizer, an AI agent that autonomously tunes vLLM serving configs, achieving a 4x throughput-latency score gain on Gemma 4 26B with zero manual tweaking.