Gemma 4 Update: Better Tool Calling
solyarisoftware · x · 2026-07-16
Google has rolled out updates to Gemma 4 aimed at local AI and agent workflows, focusing on:
- More stable tool calling: Post-fix, tool execution is more accurate and consistent.
- More complete responses: Reduces issues with premature truncation and insufficient answers.
- Faster inference: Flash Attention 4 delivers up to a 70% prefill speedup on Hopper GPUs.
- Vision and chat template optimizations: Enhances detail rendering and smooths out dialogue formats.
Driven by community feedback, this update clearly leans into local model and tool-use scenarios.
Related event: Google Rolls Out Broad Gemma 4 Improvements(7 posts)→
More from coding & agent
- FactoryAI gave back its first millions, then shipped Droid CLI two years later — matanSF · 2026-07-22
- Devin Outposts aims to run AI agents on any machine, from Mac minis to Kubernetes clusters — blaizedsouza · 2026-07-22
- Hermes Agent Refactoring Proposal: Decoupling via Event Bus and Monorepo Slicing — Promptmethus · 2026-07-22
- ty now reads Pydantic config keywords and field metadata — charliermarsh · 2026-07-22
- Pensar Launches AI Security Agent to Autonomously Discover and Patch 0-Days — andriy_mulyar · 2026-07-22
- ty adds first-class Pydantic support, including strict and lax field handling — charliermarsh · 2026-07-22