Instructions to use leon2k2k2k/qwen2.5-3b-countdown-grpo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use leon2k2k2k/qwen2.5-3b-countdown-grpo with Transformers:

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

pipe = pipeline("text-generation", model="leon2k2k2k/qwen2.5-3b-countdown-grpo")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("leon2k2k2k/qwen2.5-3b-countdown-grpo")
model = AutoModelForMultimodalLM.from_pretrained("leon2k2k2k/qwen2.5-3b-countdown-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 Settings

vLLM

How to use leon2k2k2k/qwen2.5-3b-countdown-grpo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "leon2k2k2k/qwen2.5-3b-countdown-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": "leon2k2k2k/qwen2.5-3b-countdown-grpo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/leon2k2k2k/qwen2.5-3b-countdown-grpo

SGLang

How to use leon2k2k2k/qwen2.5-3b-countdown-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 "leon2k2k2k/qwen2.5-3b-countdown-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": "leon2k2k2k/qwen2.5-3b-countdown-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 "leon2k2k2k/qwen2.5-3b-countdown-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": "leon2k2k2k/qwen2.5-3b-countdown-grpo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use leon2k2k2k/qwen2.5-3b-countdown-grpo with Docker Model Runner:
```
docker model run hf.co/leon2k2k2k/qwen2.5-3b-countdown-grpo
```

Qwen2.5-3B Countdown GRPO (iteration 300)

Qwen2.5-3B-Base trained with GRPO and a verifiable reward on the Countdown task for 300 iterations, directly from the base model with no SFT.

The headline: GRPO sharpens what the base could already do (add/sub coverage rises from 54% to 94% pass@10) but drives multiplication, a skill the base almost never produced, down to 0% (the base itself barely cleared the floor on these, about 7 correct samples in 1,500). RL reinforces existing behavior; it does not install a skill the model rarely generates on its own.

Full writeup: https://leon2k2k2k.github.io/blog/2026/grpo-sft-teaching-reasoning-through-arithmetic/ Companion: SFT-then-GRPO model | SFT dataset

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("leon2k2k2k/qwen2.5-3b-countdown-grpo")
tok = AutoTokenizer.from_pretrained("leon2k2k2k/qwen2.5-3b-countdown-grpo")

The model expects the Countdown prompt format: reason inside <think> </think>, give the final equation inside <answer> </answer>.

Results

300 held-out problems (150 add/sub, 150 needs-mult), 10 samples per problem at temperature 0.7.

cell	pass@1	pass@10
add/sub, 3 numbers	80.2%	98.9%
add/sub, 4 numbers	44.1%	85.7%
needs-mult, 3 numbers	0.0%	0.0%
needs-mult, 4 numbers	0.0%	0.0%

Training

One H100. GRPO via nano-aha-moment, following the DeepSeek-R1 recipe. G = 4, learning rate 1e-6, KL 0.001, temperature 1.0, 1024-token budget, 300 iterations. Reward = 1.0 for a well-formed <think>/<answer> response, plus 1.0 if the equation uses each number once and equals the target.

License and attribution

This is a fine-tune of Qwen2.5-3B by the Qwen team, and is released under the same Qwen Research License. The base model and its weights are their work; this repo only adds GRPO fine-tuning on Countdown.

Downloads last month: -

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for leon2k2k2k/qwen2.5-3b-countdown-grpo

Base model

Qwen/Qwen2.5-3B

Finetuned

(427)

this model

Paper for leon2k2k2k/qwen2.5-3b-countdown-grpo

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Paper • 2501.12948 • Published Jan 22, 2025 • 454