Instructions to use AxionML/Gemma-4-31B-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AxionML/Gemma-4-31B-NVFP4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="AxionML/Gemma-4-31B-NVFP4")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("AxionML/Gemma-4-31B-NVFP4")
model = AutoModelForMultimodalLM.from_pretrained("AxionML/Gemma-4-31B-NVFP4", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AxionML/Gemma-4-31B-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AxionML/Gemma-4-31B-NVFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AxionML/Gemma-4-31B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/AxionML/Gemma-4-31B-NVFP4

SGLang

How to use AxionML/Gemma-4-31B-NVFP4 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "AxionML/Gemma-4-31B-NVFP4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AxionML/Gemma-4-31B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "AxionML/Gemma-4-31B-NVFP4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AxionML/Gemma-4-31B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use AxionML/Gemma-4-31B-NVFP4 with Docker Model Runner:
```
docker model run hf.co/AxionML/Gemma-4-31B-NVFP4
```

AxionML Gemma-4-31B-NVFP4

Mirrored by AxionML for open-source serving and deployment use cases. Part of AxionML's effort to provide ready-to-serve quantized models for the community.

Quantized by NVIDIA. The weights in this repository are an unmodified copy of nvidia/Gemma-4-31B-IT-NVFP4 (revision 4135a98a9b728a548947683219633b25682223ac). All credit for the quantization belongs to NVIDIA.

This is an NVFP4-quantized version of google/gemma-4-31B-it (30.7B parameters), quantized with NVIDIA Model Optimizer.

About NVFP4 quantization: NVFP4 on Blackwell couples a compact E2M1 FP4 codebook with blockwise FP8 (E4M3) scaling over 16-element micro-blocks, so that 4-bit stored values remain numerically useful for neural-network computation. The E2M1 codebook provides a small, nonuniform set of representable magnitudes up to ±6 and relies on saturating behavior rather than IEEE NaN/Inf encodings to maximize usable range per bit. Using an FP8 block scale (rather than power-of-two-only E8M0) enables fractional scales and error-minimizing scale selection. On Blackwell Tensor Cores, native FP4 multipliers exploit E2M1 simplicity while higher-precision FP32 accumulation protects dot-product accuracy.

Ready for commercial and non-commercial use under Apache 2.0 (Gemma).

Model Summary


Architecture	Dense transformer, hybrid sliding-window + global attention, p-RoPE
Parameters	30.7B
Input	Text, image, video (as frames)
Context Length	256K tokens
Vocabulary Size	262,144
Checkpoint Size	~32.6 GB

Evaluation Results

Benchmark	BF16	NVFP4
GPQA Diamond	85.80	85.35
AIME 2025	87.92	87.60
MMLU Pro	85.25	84.94
LiveCodeBench (pass@1)	82.49	82.27
SciCode subtask acc (pass@1)	33.61	33.18
Terminal-Bench Hard (pass@1)	27.08	27.08

Scores reported by NVIDIA for this checkpoint. temperature=1.0, top_p=0.95, max_new_tokens=131072.

Quantization Details

Quantization format: NVFP4 (W4A4, group size 16) on the MLP / feed-forward linear layers; attention, embeddings, lm_head and the vision tower kept in BF16
KV cache: FP8
Calibration dataset: cnn_dailymail
Tool: NVIDIA Model Optimizer

Usage

Deploy with SGLang

python3 -m sglang.launch_server \
    --model-path AxionML/Gemma-4-31B-NVFP4 \
    --quantization modelopt_fp4 \
    --tp 1 \
    --reasoning-parser gemma4 \
    --tool-call-parser gemma4

Deploy with vLLM

vllm serve AxionML/Gemma-4-31B-NVFP4 \
    --quantization modelopt \
    --tensor-parallel-size 8

Sister checkpoint: AxionML/Gemma-4-12B-NVFP4 uses the same MLP-only recipe.

Limitations

The base model was trained on data that may contain toxic language and societal biases. The quantized model inherits these limitations. It may generate inaccurate, biased, or offensive content. Please refer to the original model card and the upstream quantized model card for full details.

Credits

Base model: google/gemma-4-31B-it
Quantization: nvidia/Gemma-4-31B-IT-NVFP4 by NVIDIA
Mirror: AxionML

Downloads last month: 599

Safetensors

Model size

21B params

Tensor type

BF16

Model tree for AxionML/Gemma-4-31B-NVFP4

Base model

google/gemma-4-31B

Finetuned

google/gemma-4-31B-it

Quantized

(342)

this model