Gemma-4-31B-it-JANG_4M

OsaurusAI/Gemma-4-31B-it-JANG_4M

Gemma-4-31B-it-JANG_4M is a image-text-to-text model from OsaurusAI. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as gemma.

<p align="center" <a href="https://osaurus.ai"<img src="https://cdn-avatars.huggingface.co/v1/production/uploads/69d00705ce8872981c6c4fce/GWKjOwezSOhW5iuKpDwq.png" alt="Osaurus AI" width="120"</a </p

Downloads · 30 days

1% of all-time downloads

All-time downloads

1.1K

Public

Parameters

39.8B

22.7 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors22.7 GB · 100%

Parameter types

How the weights are stored.

U3239.2B · 99%

Try a prompt

Base models

google/gemma-4-31b-it

Model card

From the Hugging Face model README

<a href="https://osaurus.ai"><img src="https://cdn-avatars.huggingface.co/v1/production/uploads/69d00705ce8872981c6c4fce/GWKjOwezSOhW5iuKpDwq_.png" alt="Osaurus AI" width="120"></a> <h3 align="center">Gemma 4 31B-it — JANG_4M (Mixed-Precision, 4-bit)</h3> JANG — Jang Adaptive N-bit Grading | Mixed-Precision Quantization for Apple Silicon <a href="https://osaurus.ai"><img src="https://img.shields.io/badge/Web-osaurus.ai-blue" alt="Website"></a>  <a href="https://github.com/jjang-ai/jangq"><img src="https://img.shields.io/badge/GitHub-JANG_Source-blue?logo=github" alt="GitHub"></a>  <a href="https://pypi.org/project/jang/"><img src="https://img.shields.io/pypi/v/jang?label=PyPI&color=green" alt="PyPI"></a>  <a href="https://huggingface.co/OsaurusAI"><img src="https://img.shields.io/badge/HF-OsaurusAI-yellow?logo=huggingface" alt="OsaurusAI"></a>

Osaurus natively supports JANG models. Download at osaurus.ai.

Model Details

Property	Value
Base Model	`google/gemma-4-31b-it`
Architecture	Dense Transformer + Hybrid Sliding/Global Attention
Parameters	31B (29.2B weights)
Profile	JANG_4M (CRITICAL=8-bit, COMPRESS=4-bit)
Avg Bits/Weight	5.1
Model Size	18 GB
Vision	Yes (multimodal, float16 passthrough)
Context Length	128K tokens
Layers	60
Format	JANG v2 (MLX-native safetensors, instant load)

JANG_4M Bit Allocation

Tier	Components	Bits
CRITICAL	Attention (Q/K/V/O), embeddings	8
COMPRESS	MLP (gate, up, down proj), remaining weights	4

JANG protects attention at full precision while compressing MLP weights — where dense models are most tolerant of quantization. Vision encoder is preserved in float16 for full multimodal quality.

Vision Weight Verification

All 355 vision tower tensors verified present and non-zero. The 31B dense model is text+vision (no audio tower).

Component	Tensor Count	Status
Vision Tower (SigLIP)	355	All non-zero
Language Model	remaining	All non-zero

Benchmarks

200-question MMLU (20 per subject x 10 subjects). Thinking OFF (enable_thinking=False), greedy decoding (temp=0.0).

Subject	JANG_4M
Abstract Algebra	13/20
Anatomy	13/20
Astronomy	17/20
College CS	14/20
College Physics	14/20
HS Biology	19/20
HS Chemistry	15/20
HS Mathematics	9/20
Logical Fallacies	19/20
World Religions	20/20
Total	153/200 (76.5%)

Architecture Highlights

Dense transformer with 60 layers
Hybrid attention: sliding-window + full-attention layers (every 6th layer is full)
Dual head dimensions: 256 (sliding) / 512 (global)
K=V weight sharing on global attention layers
Vision encoder preserved in float16 for multimodal inference

Usage

# Requires Osaurus (https://osaurus.ai)
osaurus serve OsaurusAI/Gemma-4-31B-it-JANG_4M

Requirements

Apple Silicon Mac with 24+ GB unified memory
Osaurus or compatible MLX inference engine with Gemma 4 support

Quantized by <a href="https://osaurus.ai">Osaurus AI</a> using <a href="https://github.com/jjang-ai/jangq">JANG</a>

Task: Image-Text-to-Text
Library: mlx
Type: gemma4
License: gemma
Languages: en
Created: Apr 6, 2026
Updated: Apr 17, 2026