Instructions to use GadflyII/Qwen3-Coder-Next-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use GadflyII/Qwen3-Coder-Next-NVFP4 with Transformers:

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

pipe = pipeline("text-generation", model="GadflyII/Qwen3-Coder-Next-NVFP4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("GadflyII/Qwen3-Coder-Next-NVFP4")
model = AutoModelForCausalLM.from_pretrained("GadflyII/Qwen3-Coder-Next-NVFP4")
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 GadflyII/Qwen3-Coder-Next-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "GadflyII/Qwen3-Coder-Next-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": "GadflyII/Qwen3-Coder-Next-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/GadflyII/Qwen3-Coder-Next-NVFP4

SGLang

How to use GadflyII/Qwen3-Coder-Next-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 "GadflyII/Qwen3-Coder-Next-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": "GadflyII/Qwen3-Coder-Next-NVFP4",
		"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 "GadflyII/Qwen3-Coder-Next-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": "GadflyII/Qwen3-Coder-Next-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use GadflyII/Qwen3-Coder-Next-NVFP4 with Docker Model Runner:
```
docker model run hf.co/GadflyII/Qwen3-Coder-Next-NVFP4
```

Model requests?

by pathosethoslogos - opened Feb 15

Discussion

pathosethoslogos

Feb 15

Do you take NVFP4 model requests?

GadflyII

Owner Feb 16

Sure, hit me

pathosethoslogos

Feb 16

mistralai/Devstral-2-123B-Instruct-2512 😊

Been using your GLM 4.7 Flash religiously

GadflyII

Owner Feb 16

I started it, but it will take awhile to finish and test.

pathosethoslogos

Feb 17

Perfect, thaaanks!

verjoz

Feb 17

Hi GadflyII,

Can I request as well an NVFP4 quantization? 😊

Your Qwen3-Coder-NEXT-NVFP4 rocks (my friends are using it), but I cannot load as low on VRAM.
https://huggingface.co/GadflyII/Qwen3-Coder-Next-NVFP4

So if you could do from the REAP (48B) version of the original 80B, I would be awesome!
(From this: https://huggingface.co/Mattepiu/Qwen3-Coder-Next-REAP-48B-A3B )

The MXFP4 version I tested and despite the 48B size still usable:
https://huggingface.co/noctrex/Qwen3-Coder-Next-REAP-48B-A3B-MXFP4_MOE-GGUF

But I would need NVFP4 to leverage from the Blackwell architecture... (2x5060 Ti 16gb)

Thank you very much in advance!

GadflyII

Owner Feb 17

I will look at the reap model, I am not sure if I can quant after they reap the weights, I may have to make nvfp4, then reap them

verjoz

Feb 18

Oh, I see, I would really appreciate if you could do that!
(please keep in mind that this should be to fitted on a 2x16Gb, so must be no more then ~25Gb 😊)
Thank you in advance.

tkg61

Feb 20

@GadflyII could you do this model? https://huggingface.co/huihui-ai/Huihui-gpt-oss-120b-BF16-abliterated

pathosethoslogos

Feb 22

@tkg61 GPT OSS 120B model is already 4bit.

tkg61

Feb 23

@pathosethoslogos i realize but the BF16 in the name is sort of confusing then…

verjoz

Feb 24

Qwen3.5 35b finially here 🚀
I think many of us would need an NVFP4 of this:
https://huggingface.co/Qwen/Qwen3.5-35B-A3B
As no one created yet a NVFP4 of this 😀

If you could do that I would really appriciate that!

GadflyII

Owner Mar 4

I have not looked at it yet, but I will.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment