Instructions to use Nanbeige/Nanbeige4-3B-Thinking-2510 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nanbeige/Nanbeige4-3B-Thinking-2510 with Transformers:

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

pipe = pipeline("text-generation", model="Nanbeige/Nanbeige4-3B-Thinking-2510")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Nanbeige/Nanbeige4-3B-Thinking-2510")
model = AutoModelForCausalLM.from_pretrained("Nanbeige/Nanbeige4-3B-Thinking-2510", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Nanbeige/Nanbeige4-3B-Thinking-2510 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Nanbeige/Nanbeige4-3B-Thinking-2510"
# 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/Nanbeige4-3B-Thinking-2510",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nanbeige/Nanbeige4-3B-Thinking-2510

SGLang

How to use Nanbeige/Nanbeige4-3B-Thinking-2510 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/Nanbeige4-3B-Thinking-2510" \
    --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/Nanbeige4-3B-Thinking-2510",
		"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/Nanbeige4-3B-Thinking-2510" \
        --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/Nanbeige4-3B-Thinking-2510",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Nanbeige/Nanbeige4-3B-Thinking-2510 with Docker Model Runner:
```
docker model run hf.co/Nanbeige/Nanbeige4-3B-Thinking-2510
```

Nanbeige4-3B-Thinking-2510

Commit History

Update README.md

78d9793
verified

flust commited on Oct 23, 2025

Update README.md

c905360
verified

flust commited on Oct 23, 2025

Update README.md

caa56a6
verified

flust commited on Oct 23, 2025

Upload performance.png

e4d1183
verified

flust commited on Oct 23, 2025

Update README.md

00ce2e9
verified

flust commited on Oct 23, 2025

Update README.md

5171cfe
verified

flust commited on Oct 23, 2025

Update README.md

a833e5a
verified

flust commited on Oct 22, 2025

Update README.md

acf0622
verified

flust commited on Oct 22, 2025

Update README.md

28ddc69
verified

flust commited on Oct 22, 2025

Update README.md

0402d3f
verified

flust commited on Oct 22, 2025

Update README.md

e7fc947
verified

flust commited on Oct 22, 2025

Update README.md

e231137
verified

flust commited on Oct 22, 2025

Update README.md

f652fe3
verified

flust commited on Oct 22, 2025

Update README.md

5351ee1
verified

flust commited on Oct 22, 2025

Upload performance.png

4828b0d
verified

leran1995 commited on Oct 22, 2025

Delete figures/performance.png

d631dd7
verified

leran1995 commited on Oct 22, 2025

Update README.md

fa537e9
verified

leran1995 commited on Oct 22, 2025

Update README.md

0e4dcca
verified

leran1995 commited on Oct 22, 2025

Delete figures/tmp

0c8cb6c
verified

leran1995 commited on Oct 22, 2025

Upload 2 files

01ec1c2
verified

leran1995 commited on Oct 22, 2025

Create figures/tmp

b8c48f9
verified

leran1995 commited on Oct 22, 2025

Update README.md

295361c
verified

leran1995 commited on Oct 22, 2025

Upload 5 files

9383050
verified

ChenZongchao commited on Oct 22, 2025

Upload pytorch_model.bin

dec4663
verified

ChenZongchao commited on Oct 22, 2025

Upload tokenizer.model

87dbf66
verified

ChenZongchao commited on Oct 22, 2025

initial commit

56e9cea
verified

leran1995 commited on Oct 22, 2025

Commit History

Update README.md 78d9793 verified

Update README.md c905360 verified

Update README.md caa56a6 verified

Upload performance.png e4d1183 verified

Update README.md 00ce2e9 verified

Update README.md 5171cfe verified

Update README.md a833e5a verified

Update README.md acf0622 verified

Update README.md 28ddc69 verified

Update README.md 0402d3f verified

Update README.md e7fc947 verified

Update README.md e231137 verified

Update README.md f652fe3 verified

Update README.md 5351ee1 verified

Upload performance.png 4828b0d verified

Delete figures/performance.png d631dd7 verified

Update README.md fa537e9 verified

Update README.md 0e4dcca verified

Delete figures/tmp 0c8cb6c verified

Upload 2 files 01ec1c2 verified

Create figures/tmp b8c48f9 verified

Update README.md 295361c verified

Upload 5 files 9383050 verified

Upload pytorch_model.bin dec4663 verified

Upload tokenizer.model 87dbf66 verified

initial commit 56e9cea verified

Update README.md

78d9793
verified

Update README.md

c905360
verified

Update README.md

caa56a6
verified

Upload performance.png

e4d1183
verified

Update README.md

00ce2e9
verified

Update README.md

5171cfe
verified

Update README.md

a833e5a
verified

Update README.md

acf0622
verified

Update README.md

28ddc69
verified

Update README.md

0402d3f
verified

Update README.md

e7fc947
verified

Update README.md

e231137
verified

Update README.md

f652fe3
verified

Update README.md

5351ee1
verified

Upload performance.png

4828b0d
verified

Delete figures/performance.png

d631dd7
verified

Update README.md

fa537e9
verified

Update README.md

0e4dcca
verified

Delete figures/tmp

0c8cb6c
verified

Upload 2 files

01ec1c2
verified

Create figures/tmp

b8c48f9
verified

Update README.md

295361c
verified

Upload 5 files

9383050
verified

Upload pytorch_model.bin

dec4663
verified

Upload tokenizer.model

87dbf66
verified

initial commit

56e9cea
verified