CoolFace
Modelpublic

AtomicChat/gemma-4-E2B-it-GGUF

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face
0likes2kdownloads
Model Card

<center>

<div style="display:flex; justify-content:center; align-items:center; gap:2%; max-width:560px; margin:0 auto;"> <a href="https://atomic.chat" style="flex:0 1 auto; min-width:0;"><img src="https://huggingface.co/AtomicChat/gemma-4-E2B-it-GGUF/resolve/main/pillatomicv3.png" alt="Atomic Chat" style="width:100%; height:auto; max-width:186px;"></a> <a href="https://discord.gg/8wGSsvmg4V" style="flex:0 1 auto; min-width:0;"><img src="https://huggingface.co/AtomicChat/gemma-4-E2B-it-GGUF/resolve/main/pilldiscordv3.png" alt="Join Discord" style="width:100%; height:auto; max-width:184px;"></a> <a href="https://github.com/AtomicBot-ai/Atomic-Chat" style="flex:0 1 auto; min-width:0;"><img src="https://huggingface.co/AtomicChat/gemma-4-E2B-it-GGUF/resolve/main/pillgithubv3.png" alt="GitHub" style="width:100%; height:auto; max-width:141px;"></a> </div>

<br/>

<img src="https://huggingface.co/AtomicChat/gemma-4-E2B-it-GGUF/resolve/main/hero.png" alt="Gemma 4 E2B" style="width:100%; max-width:100%; height:auto; margin-bottom:0.6em;"/>

<div style="display:flex; justify-content:center; gap:0.5em;"> <a href="https://huggingface.co/google/gemma-4-E2B-it"><strong>Base model: google/gemma-4-E2B-it</strong></a> </div> </center>

Gemma 4 E2B, 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

  • —2.3B effective (5.1B with embeddings) parameters: the weights this repo quantizes.
  • —Context length: 128K tokens, as published by Google.
  • —35 layers: Dense decoder, hybrid sliding-window (512) and global attention.
  • —Modalities: the base model handles Text, Image, Audio; this repo ships text-only quants, it carries no vision projector.
  • —Full imatrix ladder: every quant is calibrated with an importance matrix, published here alongside the quants.
  • —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 --jinja so the Gemma 4 E2B chat template is applied. Without it the model can emit malformed turns.

Model Overview

PropertyValue
Base modelgoogle/gemma-4-E2B-it
Parameters2.3B effective (5.1B with embeddings)
Layers35
Sliding window512 tokens
Context length128K tokens
Vocabulary262K
ModalitiesText, Image, Audio in the base model; text only in this repo, it ships no vision projector
ArchitectureDense decoder, hybrid sliding-window (512) and global attention, 8 attention heads over 1 KV head, Gemma4ForConditionalGeneration
This repoGGUF quants (imatrix); the importance matrix is published here as imatrix-coding.gguf. Quants: Q2_K, IQ3_M, Q3_K_M, Q3_K_L, IQ4_XS, Q4_K_S, Q4_K_M, Q5_K_S, Q5_K_M, Q6_K, Q8_0

Benchmarks

BenchmarkScore
MMLU Pro60.0%
AIME 2026 no tools37.5%
LiveCodeBench v644.0%
Codeforces ELO633
GPQA Diamond43.4%
Tau2 (average over 3)24.5%
BigBench Extra Hard21.9%
MMMLU67.4%
MMMU Pro44.2%
OmniDocBench 1.5 (average edit distance, lower is better)0.290
MATH-Vision52.4%
MedXPertQA MM23.5%
CoVoST33.47
FLEURS (lower is better)0.09
MRCR v2 8 needle 128k (average)19.1%

Scores are Google's published results for the base google/gemma-4-E2B-it, not our own measurements. Quantization preserves the large majority of this; Q4_K_M and up stay close to full precision.

Choosing a quant

QuantSizeNotes
Q2_K3.0 GBSmallest K-quant. Minimal RAM, clear quality drop.
IQ3_M3.1 GBBeats Q3 at a similar size thanks to imatrix. Best low-RAM pick.
Q3_K_M3.2 GBLow quality but usable.
Q3_K_L3.3 GBA step above Q3KM.
IQ4_XS3.3 GBExcellent quality for size. Recommended low-bit.
Q4_K_S3.4 GBCompact 4-bit, fast.
`Q4_K_M`3.4 GBRecommended default. Best balance of size, speed and quality.
Q5_K_S3.6 GBHigher quality, slightly more compact than Q5KM.
Q5_K_M3.6 GBHigher quality, low loss.
Q6_K3.8 GBNear lossless, noticeably lighter than Q8_0.
Q8_05.0 GBEffectively lossless, reference quality.
[!TIP] Pick the largest file that fits your (V)RAM with room for context. Q4_K_M is the sweet spot for most setups; Q6_K or Q8_0 for maximum fidelity.

Get started

Run Gemma 4 E2B locally with:

  • —[Atomic Chat](https://atomic.chat): the easiest path. Open the app, search AtomicChat/gemma-4-E2B-it-GGUF, pick a quant, hit Use this model.
  • —llama.cpp: llama-server -hf AtomicChat/gemma-4-E2B-it-GGUF:Q4_K_M --jinja -c 8192
  • —Ollama: ollama run hf.co/AtomicChat/gemma-4-E2B-it-GGUF:Q4_K_M
  • —LM Studio / Jan: search the repo id, download any quant.

Best practices

ParameterValue
temperature1.0
top_p0.95
top_k64

Google's recommended sampling configuration for google/gemma-4-E2B-it.

Run in llama.cpp

bash
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
bash
./llama.cpp/build/bin/llama-server \
    -hf AtomicChat/gemma-4-E2B-it-GGUF:Q4_K_M \
    --jinja -ngl 99 -c 8192 -fa on

How these were made

  1. 1.Download google/gemma-4-E2B-it (original weights).
  2. 2.Convert to f16 GGUF with llama.cpp.
  3. 3.Build an importance matrix over our calibration corpus, published here as imatrix-coding.gguf.
  4. 4.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.