Instructions to use DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6 with Transformers:

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

pipe = pipeline("text-generation", model="DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6")
model = AutoModelForCausalLM.from_pretrained("DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6

SGLang

How to use DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6 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 "DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6" \
    --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": "DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6" \
        --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": "DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6 with Docker Model Runner:
```
docker model run hf.co/DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-4over6
```

Add metadata and link to paper

by nielsr HF Staff - opened about 1 month ago

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+28

-2

Files changed (1) hide show

README.md +28 -2

README.md CHANGED Viewed

@@ -1,8 +1,34 @@
 ---
-license: apache-2.0
 base_model:
 - mistralai/Mistral-Nemo-Instruct-2407
 tags:
 - nvfp4
 ---
-Quantized NVFP4 version of [Mistral-Nemo-Instruct-2407](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) with [Four Over Six](https://arxiv.org/abs/2512.02010) adaptive block scaling, created to compare against my [hybrid quant]([https://huggingface.co/DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-FP8). Made with the same version of llm-compressor and compressed-tensors, using the same calibration data, to isolate the variables as much as possible.

 ---
 base_model:
 - mistralai/Mistral-Nemo-Instruct-2407
+license: apache-2.0
 tags:
 - nvfp4
+library_name: transformers
+pipeline_tag: text-generation
 ---
+Quantized NVFP4 version of [Mistral-Nemo-Instruct-2407](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) with [Four Over Six](https://arxiv.org/abs/2512.02010) adaptive block scaling, created to compare against my [hybrid quant](https://huggingface.co/DataSnake/Mistral-Nemo-Instruct-2407-NVFP4-FP8). Made with the same version of llm-compressor and compressed-tensors, using the same calibration data, to isolate the variables as much as possible.
+## Resources
+- **Paper:** [Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling](https://huggingface.co/papers/2512.02010)
+- **Repository:** [GitHub - mit-han-lab/fouroversix](https://github.com/mit-han-lab/fouroversix)
+## Method
+Four Over Six (4/6) is a modification to the block-scaled NVFP4 quantization algorithm that yields reduced quantization error. Unlike integer formats, floating point formats have non-uniform step sizes which create larger quantization error on larger values. 4/6 takes advantage of this by adaptively scaling some blocks to smaller FP4 values, making the distribution of representable values more uniform and reducing quantization error for near-maximal values.
+## Citation
+```bibtex
+@misc{cook2025sixaccuratenvfp4quantization,
+      title={Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling},
+      author={Jack Cook and Junxian Guo and Guangxuan Xiao and Yujun Lin and Song Han},
+      year={2025},
+      eprint={2512.02010},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2512.02010},
+}
+```