Instructions to use amu870/test-v2.1-grpo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use amu870/test-v2.1-grpo with Transformers:

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

pipe = pipeline("text-generation", model="amu870/test-v2.1-grpo")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("amu870/test-v2.1-grpo")
model = AutoModelForCausalLM.from_pretrained("amu870/test-v2.1-grpo")
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 amu870/test-v2.1-grpo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "amu870/test-v2.1-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": "amu870/test-v2.1-grpo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/amu870/test-v2.1-grpo

SGLang

How to use amu870/test-v2.1-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 "amu870/test-v2.1-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": "amu870/test-v2.1-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 "amu870/test-v2.1-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": "amu870/test-v2.1-grpo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use amu870/test-v2.1-grpo with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for amu870/test-v2.1-grpo to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for amu870/test-v2.1-grpo to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for amu870/test-v2.1-grpo to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="amu870/test-v2.1-grpo",
    max_seq_length=2048,
)

Docker Model Runner
How to use amu870/test-v2.1-grpo with Docker Model Runner:
```
docker model run hf.co/amu870/test-v2.1-grpo
```

test-v2.1-grpo

This model is a fine-tuned version of amu870/test-v2.1-dpo using GRPO (Group Relative Policy Optimization) via the TRL library.

Training Pipeline

SFT → DPO → GRPO (this model)

Training Objective

This model has been optimized using GRPO with a parse-based reward function that validates structured outputs (JSON, YAML, TOML, CSV, XML).

Training Configuration

Base model (DPO merged): amu870/test-v2.1-dpo
Method: GRPO (Group Relative Policy Optimization)
Epochs: 1
Learning rate: 4e-07
Beta (KL penalty): 0.05
Generations per prompt: 8
LoRA Config: r=8, alpha=16

Reward Function

The reward function validates whether the generated output can be successfully parsed as the target format:

Parse success: 1.0
Parse failure: 0.0

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "amu870/test-v2.1-grpo"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto"
)

# Test inference
prompt = "Extract the following attributes from text and output JSON..."
inputs = tokenizer.apply_chat_template(
    [{"role": "user", "content": prompt}],
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt"
).to("cuda")
outputs = model.generate(inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0]))

License

Apache 2.0. Users must follow the original base model's license terms.

Downloads last month: 1

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for amu870/test-v2.1-grpo

Base model

Qwen/Qwen3-4B-Instruct-2507

Finetuned

amu870/test-v2.1-dpo

Finetuned

(1)

this model