Gemma 4 26B A4B 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.
Highlights
- 25.2B total / 3.8B active per token parameters: the weights this repo quantizes.
- Context length: 256K tokens, as published by Google.
- 30 layers: Dense decoder, hybrid sliding-window (1024) and global attention.
- Modalities: the base model handles Text, Image; this repo ships text-only quants, it carries no vision projector.
- Full imatrix ladder: every quant is calibrated with an importance matrix.
- Reasoning: All models in the family are designed as highly capable reasoners, with configurable thinking modes.
- Diverse & Efficient Architectures: Offers Dense and Mixture-of-Experts (MoE) variants of different sizes for scalable deployment.
[!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 26B A4B It Assistant chat template is applied. Without it the model can emit malformed turns.
Model Overview
| Property | Value |
|---|---|
| Base model | google/gemma-4-26B-A4B-it-assistant |
| Parameters | 25.2B total / 3.8B active per token |
| Layers | 30 |
| Sliding window | 1024 tokens |
| Context length | 256K tokens |
| Vocabulary | 262K |
| Modalities | Text, Image in the base model; text only in this repo, it ships no vision projector |
| Architecture | Dense decoder, hybrid sliding-window (1024) and global attention, 16 attention heads over 8 KV heads, Gemma4AssistantForCausalLM |
| This repo | GGUF quants (imatrix). Quants: Q4_K_S, Q4_K_M, Q5_K_M, Q8_0, F16 |
Benchmarks
| Benchmark | Score |
|---|---|
| MMLU Pro | 82.6% |
| AIME 2026 no tools | 88.3% |
| LiveCodeBench v6 | 77.1% |
| Codeforces ELO | 1718 |
| GPQA Diamond | 82.3% |
| Tau2 (average over 3) | 68.2% |
| HLE no tools | 8.7% |
| HLE with search | 17.2% |
| BigBench Extra Hard | 64.8% |
| MMMLU | 86.3% |
| MMMU Pro | 73.8% |
| OmniDocBench 1.5 (average edit distance, lower is better) | 0.149 |
| MATH-Vision | 82.4% |
| MedXPertQA MM | 58.1% |
| MRCR v2 8 needle 128k (average) | 44.1% |
Scores are Google's published results for the base google/gemma-4-26B-A4B-it-assistant, not our own measurements. Quantization preserves the large majority of this; Q4_K_M and up stay close to full precision.
Choosing a quant
| Quant | Size | Notes |
|---|---|---|
Q4_K_S | 321 MB | Compact 4-bit, fast. |
Q4_K_M | 325 MB | Recommended default. Best balance of size, speed and quality. |
Q5_K_M | 342 MB | Higher quality, low loss. |
Q8_0 | 462 MB | Effectively lossless, reference quality. |
F16 | 0.9 GB | Unquantized reference, twice the size of Q8_0. |
[!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.
Get started
Run Gemma 4 26B A4B It Assistant locally with:
- Atomic Chat: the easiest path. Open the app, search
AtomicChat/gemma-4-26B-A4B-it-assistant-GGUF, pick a quant, hit Use this model. - llama.cpp:
llama-server -hf AtomicChat/gemma-4-26B-A4B-it-assistant-GGUF:Q4_K_M --jinja -c 8192 - Ollama:
ollama run hf.co/AtomicChat/gemma-4-26B-A4B-it-assistant-GGUF:Q4_K_M - LM Studio / Jan: search the repo id, download any quant.
Best practices
| Parameter | Value |
|---|---|
| temperature | 1.0 |
| top_p | 0.95 |
| top_k | 64 |
Google's recommended sampling configuration for google/gemma-4-26B-A4B-it-assistant.
Run in llama.cpp
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --target llama-cli llama-server
./llama.cpp/build/bin/llama-server \
-hf AtomicChat/gemma-4-26B-A4B-it-assistant-GGUF:Q4_K_M \
--jinja -ngl 99 -c 8192 -fa on
How these were made
- Download
google/gemma-4-26B-A4B-it-assistant(original weights). - Convert to f16 GGUF with llama.cpp.
- Build an importance matrix over our calibration corpus.
- Quantize the ladder with
--imatrix.
License
Original model by Google, released under the Apache 2.0 license. Full terms: Apache 2.0. Quantized by Atomic Chat.