Instructions to use Nanbeige/Nanbeige4.1-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nanbeige/Nanbeige4.1-3B with Transformers:

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

pipe = pipeline("text-generation", model="Nanbeige/Nanbeige4.1-3B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Nanbeige/Nanbeige4.1-3B")
model = AutoModelForCausalLM.from_pretrained("Nanbeige/Nanbeige4.1-3B", device_map="auto")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Nanbeige/Nanbeige4.1-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Nanbeige/Nanbeige4.1-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Nanbeige/Nanbeige4.1-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nanbeige/Nanbeige4.1-3B

SGLang

How to use Nanbeige/Nanbeige4.1-3B 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 "Nanbeige/Nanbeige4.1-3B" \
    --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": "Nanbeige/Nanbeige4.1-3B",
		"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 "Nanbeige/Nanbeige4.1-3B" \
        --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": "Nanbeige/Nanbeige4.1-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Nanbeige/Nanbeige4.1-3B with Docker Model Runner:
```
docker model run hf.co/Nanbeige/Nanbeige4.1-3B
```

Thinks, Thinks it has done, Thinks without actual execution

#31

by kempy007 - opened Feb 24

Discussion

kempy007

Feb 24

I'm trying it out via Ollama(tomng/nanbeige4.1:latest) CTX=65536 with OpenCode 1.2.10.
The Thinking is impressive, It absolutely smokes the context window, I struggle getting model to act on its thoughts, is there a prompt that will make model be more actionable. eg write the files it thought about writing.

CKATEPTb

Feb 24

•

edited Feb 24

I'm trying it out via Ollama(tomng/nanbeige4.1:latest) CTX=65536 with OpenCode 1.2.10.
The Thinking is impressive, It absolutely smokes the context window, I struggle getting model to act on its thoughts, is there a prompt that will make model be more actionable. eg write the files it thought about writing.

https://huggingface.co/Nanbeige/Nanbeige4.1-3B/discussions/19#6997a4b034b384dc490f2526

In addition to your question, I would advise waiting for version 4.2, or even waiting for the release with Muon, which is planned for version 4.5 (or 5.0, it is not known for sure).
But if you really want to, you can experiment with Nanbeige4-3B-Thinking-2511, it thinks less.
Here is an example Modelfile that you can use.

FROM /absolute/path/to/Nanbeige4-3B-Thinking-2511-(Q8_0 or Q6_K).gguf
TEMPLATE {{ .Prompt }}
RENDERER qwen3-vl-thinking
PARSER qwen3-vl-thinking
PARAMETER num_ctx 8192
PARAMETER repeat_penalty 1
PARAMETER temperature 0.6
PARAMETER top_k 20
PARAMETER top_p 0.95

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