Instructions to use Atomic-Germ/NuExtract3-4B-NPU2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Atomic-Germ/NuExtract3-4B-NPU2 with Transformers:

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

pipe = pipeline("text-generation", model="Atomic-Germ/NuExtract3-4B-NPU2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Atomic-Germ/NuExtract3-4B-NPU2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Atomic-Germ/NuExtract3-4B-NPU2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Atomic-Germ/NuExtract3-4B-NPU2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Atomic-Germ/NuExtract3-4B-NPU2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Atomic-Germ/NuExtract3-4B-NPU2

SGLang

How to use Atomic-Germ/NuExtract3-4B-NPU2 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 "Atomic-Germ/NuExtract3-4B-NPU2" \
    --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": "Atomic-Germ/NuExtract3-4B-NPU2",
		"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 "Atomic-Germ/NuExtract3-4B-NPU2" \
        --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": "Atomic-Germ/NuExtract3-4B-NPU2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Atomic-Germ/NuExtract3-4B-NPU2 with Docker Model Runner:
```
docker model run hf.co/Atomic-Germ/NuExtract3-4B-NPU2
```

NuExtract3-4B - Q4NX for FastFlowLM (AMD Ryzen AI XDNA2)

A text-extraction Qwen3.5-4B fine-tune, converted to Q4NX for FastFlowLM.

What is Q4NX?

Q4NX is FastFlowLM's native packed-quantization format - a rearranged Q4_1 layout tuned for the NPU matrix engine's tile sizes and memory access patterns. It is not a GGUF file and it does not run on llama.cpp or Ollama; it is meant exclusively for the FastFlowLM engine on AMD Ryzen AI NPUs.

Requirements

FastFlowLM >= 0.9.45 (flm CLI)
AMD Ryzen AI processor with XDNA2 (NPU2) - Strix Point / Ryzen AI 300 series or later
Linux with the XRT NPU stack installed
~16 GB of unified system memory (Q4NX weights + activations + KV cache)

Files

File	Purpose
model.q4nx	Quantized Q4NX weights
config.json	FastFlowLM model configuration
tokenizer.json	Tokenizer
tokenizer_config.json	Special tokens and chat template
chat_template.jinja	Chat template (optional)
flm-add.py	Installer script - registers this model with FastFlowLM

Install and run

This repository uses flm-add.py, a small installer that copies the model into the FastFlowLM user directory and registers the tag. It never modifies the system FastFlowLM install.

Run python3 ./flm-add.py --help for all options. Without a clone, the same command works against the repo id directly:

python3 ./flm-add.py Atomic-Germ/NuExtract3-4B-NPU2 --tag nuextract3:4b --family qwen3.5

Kernels

FastFlowLM's NPU kernels (xclbins) are closed source and are not shipped in this repository. flm-add.py links the kernels of the official qwen3.5:4b model (Qwen3.5-4B-NPU2), because this model shares the same engine family (qwen3.5) and architecture.

Serve (OpenAI-compatible)

flm serve nuextract3:4b --port 8080

curl http://127.0.0.1:8080/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -d '{"model":"nuextract3:4b","messages":[{"role":"user","content":"Hello!"}],"max_tokens":256}'

Model

Registry tag: nuextract3:4b
Engine family: qwen3.5
Kernel source: Qwen3-4B-NPU2
Context length: 32768 tokens (from config)
model.q4nx size: 4.3 GB
Base model: Qwen/Qwen3.5-4B
License: apache-2.0

Original model card

See the upstream model card for training details, benchmarks, and upstream usage. This repository only contains the Q4NX conversion for FastFlowLM.

Upstream card: numind/NuExtract3

Downloads last month: -

Collections including Atomic-Germ/NuExtract3-4B-NPU2

NPU2/Q4NX

Collection

These work with FastFlowLM • 18 items • Updated 1 day ago

GhostWriter App

Collection

Best models for the writing assistant app GhostWriter which doesn't try to write the book for you. Find at http://github.com/Atomic-Germ/GhostWriter • 8 items • Updated 1 day ago • 1