Instructions to use ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4 with Transformers:

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

pipe = pipeline("text-generation", model="ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4")
model = AutoModelForCausalLM.from_pretrained("ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4

SGLang

How to use ppetermann/Ministral-3-8B-Instruct-2512-textonly-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 "ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4 with Docker Model Runner:
```
docker model run hf.co/ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4
```

Ministral-3-8B-Instruct-2512 text-only NVFP4

This repository contains a text-only and NVFP4-quantized derivative of:

mistralai/Ministral-3-8B-Instruct-2512-BF16

What this is

The original upstream checkpoint is multimodal. This repo contains:

the text-only language model extracted from the original checkpoint
quantized to full NVFP4
saved in compressed-tensors format
adjusted for vLLM loading

I'm running this with about 32 token/s per request in vllm using the openai style api on a GB10 based machine.

The Hugging Face UI may report unusual parameter and tensor-type information for this model. That is expected for compressed NVFP4 artifacts and should not be interpreted as the original dense model having changed to a literal 5B FP32/BF16 model.

Important notes

This is not the original upstream model.
This is not multimodal. Vision/image support was removed during extraction.
This repo is intended primarily for vLLM serving.
The config.json was patched to use:

"architectures": ["MistralForCausalLM"]

Changes:

2026-03-11 first version was reluctant to call tools, calibrated new, its calling tools fine now

Downloads last month: 343

Safetensors

Model size

5B params

Tensor type

F32

BF16

F8_E4M3

Model tree for ppetermann/Ministral-3-8B-Instruct-2512-textonly-NVFP4

Base model

mistralai/Ministral-3-8B-Base-2512

Finetuned

mistralai/Ministral-3-8B-Instruct-2512-BF16

Quantized

(17)

this model