Instructions to use tomhu/RL4TG-Qwen3-8B-OPD-GRPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tomhu/RL4TG-Qwen3-8B-OPD-GRPO with Transformers:

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

pipe = pipeline("text-generation", model="tomhu/RL4TG-Qwen3-8B-OPD-GRPO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tomhu/RL4TG-Qwen3-8B-OPD-GRPO")
model = AutoModelForCausalLM.from_pretrained("tomhu/RL4TG-Qwen3-8B-OPD-GRPO", 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 tomhu/RL4TG-Qwen3-8B-OPD-GRPO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tomhu/RL4TG-Qwen3-8B-OPD-GRPO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tomhu/RL4TG-Qwen3-8B-OPD-GRPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tomhu/RL4TG-Qwen3-8B-OPD-GRPO

SGLang

How to use tomhu/RL4TG-Qwen3-8B-OPD-GRPO 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 "tomhu/RL4TG-Qwen3-8B-OPD-GRPO" \
    --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": "tomhu/RL4TG-Qwen3-8B-OPD-GRPO",
		"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 "tomhu/RL4TG-Qwen3-8B-OPD-GRPO" \
        --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": "tomhu/RL4TG-Qwen3-8B-OPD-GRPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tomhu/RL4TG-Qwen3-8B-OPD-GRPO with Docker Model Runner:
```
docker model run hf.co/tomhu/RL4TG-Qwen3-8B-OPD-GRPO
```

RL4TG Qwen3-8B OPD GRPO

This model is a Qwen3-8B checkpoint trained for Java unit-test generation with offline policy distillation (OPD), followed by mutation-aware GRPO. It is a selected release state from the GRPO run. The repository name intentionally does not encode an internal checkpoint number.

Training configuration

Stage 1: offline policy distillation

Setting	Value
Student	`Qwen/Qwen3-8B`
Teacher	`Qwen/Qwen3-14B`
Objective	forward KL over teacher top-k logits
Teacher top-k	256
Policy-gradient term	disabled
Precision	BF16
Learning rate	`3e-6`, constant schedule
Epochs	1
Global batch size	32
Per-GPU micro batch	1
Gradient accumulation	32
Rollout samples	1
Temperature	1.0
Maximum prompt/response length	4096 / 4096 tokens
Seed	42

Stage 2: mutation-aware GRPO

Setting	Value
Starting policy	OPD model from stage 1
Training examples	1,582 mutation-applicable Defects4J training samples
Algorithm	GRPO, no KL term in reward
Configured epochs	2
Learning rate	`1e-6`
Global prompt batch	32
Rollouts per prompt	8
PPO prompt groups / expanded mini batch	4 / 32
Per-GPU micro batch	2
Gradient accumulation	8
Temperature	1.0
Entropy coefficient	`0.01`
Maximum prompt/response length	4096 / 4096 tokens
Data shuffle / seed	enabled / 42
Chat template	enabled; Qwen thinking disabled

The reward is -1 for an invalid generated test. A valid test receives 0.4 * line_coverage + 0.6 * branch_coverage + mutation_score. Mutation analysis samples up to three eligible failing mutants per rollout, with a mutation weight of 1.0. Coverage credit is gated by focal-call, assertion, and local focal-data provenance checks.

Intended use

The model is intended for research on Java unit-test generation and reinforcement learning for software engineering. Generated tests must be compiled and executed in an isolated environment before use.