Gemma 4 31B Instruction-Tuned

Google DeepMind🇺🇸 United States
active
Context window262K tokens

Version History

4.0major

Gemma 4 introduces multimodal capabilities (text, image, video, audio), extended context windows up to 256K tokens, native reasoning modes, function calling support, and both dense and MoE architectures optimized for deployment from edge devices to servers.

Coverage

model release

Google releases Gemma 4 family with 31B model, 256K context, multimodal capabilities

Google DeepMind released the Gemma 4 family of open-weights models ranging from 2.3B to 31B parameters, featuring up to 256K token context windows and native support for text, image, video, and audio inputs. The flagship 31B model scores 85.2% on MMLU Pro and 89.2% on AIME 2026, with a smaller 26B MoE variant requiring only 3.8B active parameters for faster inference.

2 min read