Instructions to use Nanbeige/Nanbeige4.3-33B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nanbeige/Nanbeige4.3-33B with Transformers:

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

pipe = pipeline("text-generation", model="Nanbeige/Nanbeige4.3-33B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Nanbeige/Nanbeige4.3-33B")
model = AutoModelForCausalLM.from_pretrained("Nanbeige/Nanbeige4.3-33B", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Nanbeige/Nanbeige4.3-33B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Nanbeige/Nanbeige4.3-33B"
# 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/Nanbeige4.3-33B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nanbeige/Nanbeige4.3-33B

SGLang

How to use Nanbeige/Nanbeige4.3-33B 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/Nanbeige4.3-33B" \
    --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/Nanbeige4.3-33B",
		"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/Nanbeige4.3-33B" \
        --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/Nanbeige4.3-33B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Nanbeige/Nanbeige4.3-33B with Docker Model Runner:
```
docker model run hf.co/Nanbeige/Nanbeige4.3-33B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

1. Introduction

In Nanbeige4.3, we take a foundational step toward pushing the boundaries of lightweight agents. Our strategy centers on building a high-capacity teacher agent first, which will subsequently anchor the distillation of advanced agentic capabilities into smaller models.

To this end, we first introduce Nanbeige4.3-33B, designed as the flagship teacher model for the Nanbeige4.3 series. Delivering state-of-the-art agentic performance among open models in the ~30B parameter class, Nanbeige4.3-33B excels at complex coding and professional cowork tasks. It not only offers a powerful, cost-effective option for immediate real-world deployment, but also lays the necessary foundation to power our upcoming small model distillation pipeline (e.g., for Nanbeige4.3-3B).

2. Model Performance

We compare Nanbeige4.3-33B with recently released frontier models at around the 30B-parameter scale: Qwen3.8-27B, Muse-Glimmer-30B, and Gemma4-31B. All models are evaluated at their highest thinking effort. For each benchmark, we use the same inference hyperparameters and harness across models.

Category	Benchmark	Nanbeige4.3-33B¹	Qwen3.8-27B	Muse-Glimmer-30B	Gemma4-31B
Agentic Coding	Terminal-Bench 2.1²	80.9	73.0	61.8	47.2
	Terminal-Bench 4.0³	23.8	6.3	1.6	0.0
	DeepSWE³	65.5	42.5	4.4	2.7
	SWE-Bench Pro³	60.3	61.7	36.9	51.2
	SWE-Atlas-QnA⁴	50.8	41.9	17.7	3.3
	NL2Repo⁵	53.0	48.3	27.4	33.8
	Scicode	52.6	46.7	44.9	45.5
	SkillsBench	61.1	56.4	39.7	28.9
Agentic Cowork	Automation Bench⁶	54.2	38.3	15.2	9.7
	JobBench⁷	44.8	34.5	16.6	9.0
	GDPval⁸	84.8	86.4	75.8	71.4
	Claw-Eval_pass^3⁹	78.3	75.2	57.3	44.2
	Toolathlon-Verified¹⁰	70.4	68.5	47.2	27.8
	DeepPlanning¹²	63.3	49.2	21.4	19.4
	EnterpriseOps-Gym¹²	41.0	40.2	30.5	27.3
	YC-bench¹¹	$1,025,437	$641,290	$513,978	$265,152
	Tau3-Bench¹²	48.8	48.5	33.7	18.6
Reasoning	GPQA Diamond	89.5	89.2	83.5	91.3
	AIME 2026	95.2	94.5	94.7	89.2
	AA-LCR	84.0	82.0	83.3	72.0

¹ Nanbeige4.3-33B is evaluated with preserve_thinking=true in the chat template.
² Terminal-Bench 2.1 is evaluated using Terminus 2.
³ Terminal-Bench 4 and DeepSWE are evaluated using mini-swe-agent v2.4.6.
⁴ SWE-Atlas-QnA is evaluated using mini-swe-agent v2.4.6. Answers are evaluated against 1,304 task-specific rubrics using GPT-6 Sol as the judge, with medium reasoning effort.
⁵ NL2Repo is evaluated using Claude Code.
⁶ Automation Bench v1.0.6 includes 600 publicly released tasks.
⁷ JobBench is evaluated with the Claude Code harness equipped with a search tool.
⁸ GDPval is scored using revised rubrics.
⁹ Claw-Eval is evaluated exclusively on the Claw-Eval General subset (comprising 157 single-turn user interaction tasks).
¹⁰ Toolathlon-Verified sets no max_tokens limit for individual model requests.
¹¹ YC-bench retains only the most recent 10 turns of user dialogue as historical context before each model call.
¹² DeepPlanning, EnterpriseOps-Gym, Tau3-Bench are evaluated using their official repositories.

Nanbeige4.3-33B leads on most of the benchmarks, showing strong performance in both agentic coding and cowork tasks. Beyond its breadth across benchmarks, Nanbeige4.3-33B shows strong performance on agent tasks across a wide range of professions and industries. For example, on JobBench, it outperforms the baseline in most of the 35 occupations evaluated.

3. Quickstart

The tokenizer provides a configurable chat template for reasoning and tool-use scenarios:

enable_thinking controls whether the model generates reasoning for the current response. It is enabled by default; set it to False for non-thinking mode.
preserve_thinking controls whether reasoning from previous assistant turns is retained in a multi-turn conversation. We recommend False for general chat and question answering, and True for multi-turn tool use, office tasks, and code-agent workflows.
Passing tools enables the tool-use template. We recommend tool_call_format="xml" for the best tool-calling performance; json is also supported for compatibility.

The model supports a context length of up to 524,288 tokens (512K).

We recommend adjusting the inference settings according to the target scenario:

Scenario	Temperature	Top_p	Max New Tokens
Agentic and tool-use tasks	1.0	0.95	65,536
Reasoning and chat tasks	0.6	0.95	262,144

You can run the model with Transformers, SGLang, or vLLM:

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = 'Nanbeige/Nanbeige4.3-33B'
tokenizer = AutoTokenizer.from_pretrained(
  model_id,
  use_fast=False,
  trust_remote_code=True
)
model = AutoModelForCausalLM.from_pretrained(
  model_id,
  torch_dtype='auto',
  device_map='auto',
  trust_remote_code=True
)

messages = [{'role': 'user', 'content': 'Which number is bigger, 9.11 or 9.8?'}]
inputs = tokenizer.apply_chat_template(
  messages,
  add_generation_prompt=True,
  return_tensors='pt'
).to(model.device)
output_ids = model.generate(inputs, max_new_tokens=1024)
response = tokenizer.decode(output_ids[0][inputs.shape[-1]:], skip_special_tokens=True)
print(response)

SGLang

MODEL_PATH=/path/to/your/Nanbeige4.3-33B
python -m sglang.launch_server \
  --model-path ${MODEL_PATH} \
  --host 0.0.0.0 \
  --port 8000 \
  --tp-size 4 \
  --mem-fraction-static 0.8 \
  --reasoning-parser qwen3 \
  --tool-call-parser qwen3_coder

vLLM

MODEL_PATH=/path/to/your/Nanbeige4.3-33B
vllm serve ${MODEL_PATH} \
  --host 0.0.0.0 \
  --port 8000 \
  --tensor-parallel-size 4 \
  --gpu-memory-utilization 0.8 \
  --enable-auto-tool-choice \
  --reasoning-parser qwen3 \
  --tool-call-parser qwen3_coder

4. Contact

If you have any questions, please open an issue in this repository or contact us at nanbeige@kanzhun.com.

We also welcome collaborations to build domain-specific and industry-focused agents, including joint work on data and training.

5. Citation

If you find our model useful or would like to use it in your own work, please cite this Hugging Face project:

@misc{nanbeige43chat2026,
  title        = {Nanbeige4.3-33B},
  author       = {Nanbeige Team},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/Nanbeige/Nanbeige4.3-33B}}
}