Instructions to use Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B with Transformers:

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

pipe = pipeline("text-generation", model="Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B")
model = AutoModelForCausalLM.from_pretrained("Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B")
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 Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B

SGLang

How to use Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B 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 "Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B" \
    --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": "Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B",
		"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 "Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B" \
        --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": "Jasaxion/MathSmith-Hard-Problem-Synthesizer-Qwen3-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

MathSmith-Hard-Problem-Synthesizer-Qwen3-8B

MathSmith: Towards Extremely Hard Mathematical Reasoning by Forging Synthetic Problems with a Reinforced Policy

Overview

MathSmith is a framework for synthesizing challenging mathematical problems to enhance LLM reasoning. Rather than modifying existing problems, MathSmith constructs new ones from scratch by randomly sampling concept-explanation pairs from PlanetMath.

The model generates <rationale>–<problem> pairs, where:

<rationale>: structured reasoning describing concept integration and difficulty design.
<problem>: a single Olympiad-level mathematical question that admits a verifiable numeric or symbolic answer.

Compared with MathSmith-HC (complexity + consistency reward), MathSmith-Hard removes the consistency term to emphasize maximum reasoning depth and difficulty.

MathSmith Pipeline

The MathSmith framework consists of four main stages:

Concept Collection: Randomly sample concept–explanation pairs from PlanetMath to ensure data independence.
Supervised Fine-tuning (SFT): Train the model on collected concept–explanation pairs to establish foundational understanding.
Reinforcement Learning (RL): Optimize the model using GRPO with rewards based on:
- Structural validity
- Reasoning complexity
- Answer consistency
Weakness-Focused Self-Improvement: Iteratively identify and address model weaknesses by generating targeted problem variants.

Dependence

Transformers 4.52.4
Pytorch 2.7.0+cu126
Datasets 3.6.0
Tokenizers 0.21.1

Citation

If you find this work useful, please cite:

@article{zhan2025mathsmith,
  title={MathSmith: Towards Extremely Hard Mathematical Reasoning by Forging Synthetic Problems with a Reinforced Policy},
  author={Zhan, Shaoxiong and Lai, Yanlin and Lu, Ziyu and Lin, Dahua and Yang, Ziqing and Tan, Fei},
  journal={arXiv preprint arXiv:2508.05592},
  year={2025}
}