Skip to content
Friday, 20 November 2026
Tenesys AI News
Subscribe

Gemma 4 E2B

Change history

  1. 19 Nov 2026
    Optimizing Multi-Turn AI Agents via Tunix and Gemma 4 Reinforcement Learning
    ParameterBeforeNow
    Win rate (Plane Strike game)9.4%95%
    Average game length22.44 strikes~16 strikes
    Invalid action rateNot specified~0%

News

Models· Important· 🧪 Worth Testing

Optimizing Multi-Turn AI Agents via Tunix and Gemma 4 Reinforcement Learning

Google has expanded its Tunix framework to enable efficient multi-turn reinforcement learning on Cloud TPU v6e-8. By applying agentic Group Relative Policy Optimization (GRPO) to Gemma 4, developers can now train models that excel at spatial reasoning and iterative tool usage.

Google Developers
Gemma 4 E2B · TENESYS AI NEWS