178 likesNew · 4d old

Gemma 4-31B IT model gets new GGUF quantizations

Model · mradermacher/gemma-4-31B-it-ARA-heretic-GGUF

InferenceLLMs

What is gemma-4-31B-it-ARA-heretic-GGUF?

A collection of GGUF-formatted quantized checkpoints for the Gemma 4-31B IT-ARA-heretic model is now public. The files range from 2-bit to 8-bit, shrinking the model to as little as 0.9 GB for low-memory inference.

Who it is for

Developers who want a 31-billion-parameter LLM but are limited by GPU RAM can grab a smaller file and run it on a consumer-grade setup.

How to try it

Download a Q4_K_S file and load it with a GGUF-compatible runtime like llama.cpp

The morning email · free

Get tomorrow's biggest AI launches before everyone else.

One short email each morning: what launched, why it matters, how to try it. Five minutes, no spam, unsubscribe any time.

Email addressSubscribe

The story so far

Early Signal first spotted mradermacher/gemma-4-31B-it-ARA-heretic-GGUF on November 10, 2026, with 178 likes. On its busiest day it gained 44 likes. The project is 4 days old. Early Signal files it under Inference, LLMs.

Questions about gemma-4-31B-it-ARA-heretic-GGUF

How do I try gemma-4-31B-it-ARA-heretic-GGUF?

Download a Q4_K_S file and load it with a GGUF-compatible runtime like llama.cpp

What do I need to run gemma-4-31B-it-ARA-heretic-GGUF?

A GGUF-compatible inference engine (e.g., llama.cpp)