Qwen
Qwen/Qwen3-4B-Thinking-2507-FP8
This repo contains the FP8 version of Qwen3-4B-Thinking-2507, which has the following features: - Type: Causal Language Models - Training St...
Model Documentation
Qwen3-4B-Thinking-2507-FP8
Highlights
Over the past three months, we have continued to scale the thinking capability of Qwen3-4B, improving both the quality and depth of reasoning. We are pleased to introduce Qwen3-4B-Thinking-2507-FP8, featuring the following key enhancements:
NOTE: This version has an increased thinking length. We strongly recommend its use in highly complex reasoning tasks.
Model Overview
This repo contains the FP8 version of Qwen3-4B-Thinking-2507, which has the following features:
NOTE: This model supports only thinking mode. Meanwhile, specifying
enable_thinking=True is no longer required.Additionally, to enforce model thinking, the default chat template automatically includes
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation.
Performance
| | Qwen3-30B-A3B Thinking | Qwen3-4B Thinking | Qwen3-4B-Thinking-2507 | |--
$ For reproducibility, we report the win rates evaluated by GPT-4.1.
\& For highly challenging tasks (including PolyMATH and all reasoning and coding tasks), we use an output length of 81,920 tokens. For all other tasks, we set the output length to 32,768.
Quickstart
The code of Qwen3 has been in the latest Hugging Face
transformers and we advise you to use the latest version of transformers.With
transformers<4.51.0, you will encounter the following error:
KeyError: 'qwen3'
The following contains a code snippet illustrating how to use the model generate content based on given inputs.
python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen3-4B-Thinking-2507-FP8"
load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
conduct text completion
generated_ids = model.generate(
**model_inputs,
max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
parsing thinking content
try:
rindex finding 151668 ()
index = len(output_ids) output_ids[::-1].index(151668)
except ValueError:
index = 0
thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
print("thinking content:", thinking_content) no opening tag
print("content:", content)
For deployment, you can use
sglang>=0.4.6.post1 or vllm>=0.8.5 or to create an OpenAI-compatible API endpoint:
shell
python -m sglang.launch_server --model-path Qwen/Qwen3-4B-Thinking-2507-FP8 --context-length 262144 --reasoning-parser deepseek-r1
shell
vllm serve Qwen/Qwen3-4B-Thinking-2507-FP8 --max-model-len 262144 --enable-reasoning --reasoning-parser deepseek_r1
Note: If you encounter out-of-memory (OOM) issues, you may consider reducing the context length to a smaller value. However, since the model may require longer token sequences for reasoning, we strongly recommend using a context length greater than 131,072 when possible.
For local use, applications such as Ollama, LMStudio, MLX-LM, llama.cpp, and KTransformers have also supported Qwen3.
Note on FP8
For convenience and performance, we have provided
fp8-quantized model checkpoint for Qwen3, whose name ends with -FP8. The quantization method is fine-grained fp8 quantization with block size of 128. You can find more details in the quantization_config field in config.json.You can use the Qwen3-4B-Thinking-2507-FP8 model with serveral inference frameworks, including
transformers, sglang, and vllm, as the original bfloat16 model.Agentic Use
Qwen3 excels in tool calling capabilities. We recommend using Qwen-Agent to make the best use of agentic ability of Qwen3. Qwen-Agent encapsulates tool-calling templates and tool-calling parsers internally, greatly reducing coding complexity.
To define the available tools, you can use the MCP configuration file, use the integrated tool of Qwen-Agent, or integrate other tools by yourself.
python
from qwen_agent.agents import Assistant
Define LLM
Using OpenAI-compatible API endpoint. It is recommended to disable the reasoning and the tool call parsing
functionality of the deployment frameworks and let Qwen-Agent automate the related operations. For example,
VLLM_USE_MODELSCOPE=true vllm serve Qwen/Qwen3-4B-Thinking-2507-FP8 --served-model-name Qwen3-4B-Thinking-2507 --max-model-len 262144.
llm_cfg = {
'model': 'Qwen3-4B-Thinking-2507',
Use a custom endpoint compatible with OpenAI API:
'model_server': 'http://localhost:8000/v1', api_base without reasoning and tool call parsing
'api_key': 'EMPTY',
'generate_cfg': {
'thought_in_content': True,
},
}
Define Tools
tools = [
{'mcpServers': { You can specify the MCP configuration file
'time': {
'command': 'uvx',
'args': ['mcp-server-time', '--local-timezone=Asia/Shanghai']
},
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
},
'code_interpreter', Built-in tools
]
Define Agent
bot = Assistant(llm=llm_cfg, function_list=tools)
Streaming generation
messages = [{'role': 'user', 'content': 'https://qwenlm.github.io/blog/ Introduce the latest developments of Qwen'}]
for responses in bot.run(messages=messages):
pass
print(responses)
Best Practices
To achieve optimal performance, we recommend the following settings:
1. Sampling Parameters:
Temperature=0.6, TopP=0.95, TopK=20, and MinP=0.presence_penalty parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.2. Adequate Output Length: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 81,920 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
3. Standardize Output Format: We recommend using prompts to standardize model outputs when benchmarking.
answer field with only the choice letter, e.g., "answer": "C"."4. No Thinking Content in History: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
Citation
If you find our work helpful, feel free to give us a cite.
@misc{qwen3technicalreport,
title={Qwen3 Technical Report},
author={Qwen Team},
year={2025},
eprint={2505.09388},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2505.09388},
}
Files & Weights
| Filename | Size | Action |
|---|---|---|
| model.safetensors | 4.83 GB |