Google updates Gemma 4 chat templates with tool calling fixes, Flash Attention 4

Google released updates to Gemma 4's chat templates, fixing tool calling and reducing model laziness, and enabling Flash Attention 4 on Hopper GPUs. An interactive guide for improving Gemma 4's vision capabilities is also available.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Skeptic shares first impressions of ChatGPT Plus
- Hard sci-fi authors largely oppose LLMs, survey finds
- Anthropic tests folderless Claude Code sessions on Desktop and iOS
- Ethan Mollick: Using weaker AI for human-facing content may soon be disrespectful
- Pocket TTS training stack open-sourced