Instructions to use Nanbeige/Nanbeige4-3B-Thinking-2511 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nanbeige/Nanbeige4-3B-Thinking-2511 with Transformers:

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

pipe = pipeline("text-generation", model="Nanbeige/Nanbeige4-3B-Thinking-2511")
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-2511")
model = AutoModelForCausalLM.from_pretrained("Nanbeige/Nanbeige4-3B-Thinking-2511")
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

vLLM

How to use Nanbeige/Nanbeige4-3B-Thinking-2511 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-2511"
# 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-2511",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nanbeige/Nanbeige4-3B-Thinking-2511

SGLang

How to use Nanbeige/Nanbeige4-3B-Thinking-2511 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-2511" \
    --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-2511",
		"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-2511" \
        --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-2511",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Nanbeige/Nanbeige4-3B-Thinking-2511 with Docker Model Runner:
```
docker model run hf.co/Nanbeige/Nanbeige4-3B-Thinking-2511
```

leran1995 commited on Dec 1, 2025

Commit

4367424

verified ·

1 Parent(s): 3aa07cc

Update README.md

Browse files

Files changed (1) hide show

README.md +7 -1

README.md CHANGED Viewed

@@ -23,11 +23,17 @@ tags:
 # Introduction
 Nanbeige4-3B-Thinking-2511 is an enhanced iteration over our previous Nanbeige4-3B-Thinking-2510.
 Through advanced distillation techniques and reinforcement learning (RL) optimization, we have effectively scaled the model’s reasoning capacity, resulting in superior performance across a broad range of benchmarks.
-Notably, Nanbeige4-3B-Thinking-2511 achieves state-of-the-art (SOTA) results among models smaller than 32B parameters on Arena-Hard-V2 and BFCL-V4.
 This marks a major milestone in delivering powerful, efficient reasoning performance at a compact scale.
 <div align="center">
 <img src="figures/performance_2511.png">
 </div>

 # Introduction
 Nanbeige4-3B-Thinking-2511 is an enhanced iteration over our previous Nanbeige4-3B-Thinking-2510.
 Through advanced distillation techniques and reinforcement learning (RL) optimization, we have effectively scaled the model’s reasoning capacity, resulting in superior performance across a broad range of benchmarks.
+On math and science reasoning benchmarks, Nanbeige4-3B-Thinking-2511 outperforms Qwen3-4B-Thinking-2507, Qwen3-8B-Thinking-2504, and Qwen3-14B-Thinking-2504 with a significant margin.
+Besides, Nanbeige4-3B-Thinking-2511 achieves state-of-the-art (SOTA) results among models smaller than 32B parameters on Arena-Hard-V2 and BFCL-V4.
 This marks a major milestone in delivering powerful, efficient reasoning performance at a compact scale.
 <div align="center">
+<img src="figures/performance_reasoning.png">
+</div>
+<div align="center">
 <img src="figures/performance_2511.png">
 </div>