Instructions to use Julian2002/PDP-Qwen3-8B-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Julian2002/PDP-Qwen3-8B-SFT with Transformers:

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

pipe = pipeline("text-generation", model="Julian2002/PDP-Qwen3-8B-SFT")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Julian2002/PDP-Qwen3-8B-SFT")
model = AutoModelForCausalLM.from_pretrained("Julian2002/PDP-Qwen3-8B-SFT")
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 Julian2002/PDP-Qwen3-8B-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Julian2002/PDP-Qwen3-8B-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Julian2002/PDP-Qwen3-8B-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Julian2002/PDP-Qwen3-8B-SFT

SGLang

How to use Julian2002/PDP-Qwen3-8B-SFT 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 "Julian2002/PDP-Qwen3-8B-SFT" \
    --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": "Julian2002/PDP-Qwen3-8B-SFT",
		"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 "Julian2002/PDP-Qwen3-8B-SFT" \
        --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": "Julian2002/PDP-Qwen3-8B-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Julian2002/PDP-Qwen3-8B-SFT with Docker Model Runner:
```
docker model run hf.co/Julian2002/PDP-Qwen3-8B-SFT
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

PDP Qwen3-8B RQ3 SFT

Model Summary

This repository contains an RQ3 supervised fine-tuning checkpoint for prosecution decision prediction (PDP) experiments based on Qwen/Qwen3-8B.

Repository: Julian2002/PDP-Qwen3-8B-SFT
Source directory: checkpoint-50
Upload mode: inference_only
Dataset: PDP-Bench / pdp2k_rq3_sft

Task

Given suspect information, procedural information, and factual information, the model is trained to generate structured prosecutorial reasoning and a final decision in the PDP setting.

The target decision space contains four labels:

起诉
相对不起诉
法定不起诉
存疑不起诉

Training Output Format

The supervised target follows the project format with a reasoning block and an answer block:

<think>
...
</think>

<answer>
【适用法条】
...

【审查分析】
...

【最终结论】
决定：...
</answer>

Notes

This repository was uploaded from an intermediate training checkpoint directory. If the source was a DeepSpeed ZeRO checkpoint, additional conversion or consolidation may be required for direct standalone inference.

This model card was generated automatically by scripts/upload_RQ3_sft_checkpoint_to_hf.sh.

Downloads last month: 549

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for Julian2002/PDP-Qwen3-8B-SFT

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Finetuned

(1930)

this model

Quantizations

1 model