Gemma 4 31B (free)
A
AI Socratic
Google released Gemma 4 31B (free) on July 24, 2026. The registry currently records it as open weights.
Features a 256K token context window, configurable thinking/reasoning mode, native function calling, and multilingual support across 140+ languages. Strong on coding, reasoning, and document understanding tasks. Apache 2.0 license.
The structured profile below stays connected to the canonical model record, including its architecture, parameters, papers, benchmarks, media, licence, and source links as those fields are enriched.
Model publication
ReadyArt
Gemma 4 31B (free)Open the source for Gemma 4 31B (free)
Open weights
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function calling, and multilingual support across 140+ languages. Strong on coding, reasoning, and document understanding tasks. Apache 2.0 license.
- Licence
- apache-2.0
- Architecture
- dense
- Parameters
- 32.7B
- Context
- 262.1K tokens
Benchmarks
No benchmark observation recorded for this model.
Weightshuggingface.co/google/gemma-4-31B-itHugging Facehuggingface.coHugging Facehuggingface.coOpenRouteropenrouter.ai5 artifacts in the registry, including third-party rebuilds
About the Authors
A