No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from AtomicChat/gemma-4-31B-it-assistant-GGUF
All models by AtomicChat →
Gemma 4 31B It Assistant, self-quantized to GGUF by Atomic Chat. Built straight from Google's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline.
[!NOTE] These GGUFs are self-quantized from the original weights, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.
[!IMPORTANT] Always pass
--jinjaso the Gemma 4 31B It Assistant chat template is applied. Without it the model can emit malformed turns.
Scores are Google's published results for the base google/gemma-4-31B-it-assistant, not our own measurements. Quantization preserves the large majority of this; Q4_K_M and up stay close to full precision.
[!TIP] Pick the largest file that fits your (V)RAM with room for context.
Q4_K_Mis the sweet spot for most setups;Q6_KorQ8_0for maximum fidelity.
Run Gemma 4 31B It Assistant locally with:
AtomicChat/gemma-4-31B-it-assistant-GGUF, pick a quant, hit Use this model.llama-server -hf AtomicChat/gemma-4-31B-it-assistant-GGUF:Q4_K_M --jinja -c 8192ollama run hf.co/AtomicChat/gemma-4-31B-it-assistant-GGUF:Q4_K_MGoogle's recommended sampling configuration for google/gemma-4-31B-it-assistant.
google/gemma-4-31B-it-assistant (original weights).--imatrix.Original model by Google, released under the Apache 2.0 license. Full terms: Apache 2.0. Quantized by Atomic Chat.
git clone https://github.com/ggml-org/llama.cppcmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ONcmake --build llama.cpp/build --config Release -j --target llama-cli llama-server./llama.cpp/build/bin/llama-server \ -hf AtomicChat/gemma-4-31B-it-assistant-GGUF:Q4_K_M \ --jinja -ngl 99 -c 8192 -fa on