Gemma-4-31B available for inference on HuggingChat
July 1, 2026
HuggingChat now supports inference for the gemma-4-31B model at 1x speed. Users can test the model's performance and responsiveness directly through the Hugging Face chat interface.
HOW THIS AFFECTS YOU
●
builderYou can now benchmark and prototype with the 31B parameter version of Gemma-4 via HuggingChat.
●
researcherThis provides a direct interface to evaluate model reasoning and instruction following for this specific parameter scale.