KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#typescript #ai-agents #developer-tools #cli #codex #claude-code #dsh-plugin #mcp #react #agent #dsh #deepseek-harness

DeepMath

★ 45 Open GitHub ↗

Simple framework for training and evaluating math reasoning agents using local models, GRPO and vLLM.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# DeepMath: A Lightweight Math Reasoning Agent for LLMs

Large language models (LLMs) have made impressive strides in reasoning tasks, yet mathematical problem-solving remains a challenge. Traditional "chain-of-thought" reasoning often produces verbose explanations and error-prone arithmetic. **DeepMath** tackles this by combining a small Python executor with a fine-tuned LLM, enabling concise, computation-driven reasoning.

## DeepMath

DeepMath is built on **Qwen3-4B Thinking** and fine-tuned with **GRPO (Group Relative Policy Optimization)**. Instead of verbose text, the model emits **tiny Python snippets** for intermediate steps, runs them in a secure sandbox, and folds the results back into its reasoning, reducing errors and output length.

✅ No file I/O, no network calls, strict timeouts.

✅ Safe, deterministic, and auditable.

We evaluate DeepMath on four math datasets: **MATH500, AIME, HMMT, and HLE,** and show that:

- The math agent alone improves accuracy and reduces verbosity.

- GRPO training alone biases outputs toward brevity and correctness.

- Combining the agent with GRPO yields the largest gains.

Check out:

- 🔗 Blog: https://huggingface.co/blog/intel-deepmath

- 👉 Model: <https://huggingface.co/Intel/deepmath-v1>

## Why DeepMath?

LLMs often struggle with numeric precision and produce unnecessarily long reasoning chains. Two opportunities stand out:

1. **Offload deterministic computation** to a safe executor.

2. **Train models to prefer concise, computation-oriented traces** over verbose text.

DeepMath implements both. The model learns to generate short Python snippets, which are executed in a sandbox and reintegrated into the context. GRPO fine-tuning encourages this behavior by rewarding correctness and encouraging shorter outputs.

## How It Works

- Base model: [Qwen3-4B Thinking](https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507).
- Executor constraints: sandboxed environment, allow-list of imported modules, per-snippet timeout.
- Inference: based on [SmolAgents](https://github.com/huggingface/smolagents/), a math agent was created. vLLM is used as the inference engine.
- Training: based on the GRPO trainer in [TRL](https://github.com/huggingface/trl), we modified TRL's vLLM client and server to generate GRPO completions using our DeepMath agent.

<div align="center">
<img src="assets/trl-grpo-vllm-deepmath.png" width=600 alt="Changes to vLLM client and server in TRL library." />
</div><br>
<em>Figure 1: The vLLM client and server were modified to use the DeepMath agent in generating the candidates, while using the vLLM backend.</em>

- **Agent Interface:** During inference, the model can output normal tokens or special agent calls containing Python snippets.

- **Execution:** Snippets run in a sandboxed environment with strict safety constraints (no file I/O, no network, timeouts).

- **Design Goals:**

- **Concision:** Replace multi-line textual calculations with short, focused snippets.

- **Determinism & Safety:** Enforce strict execution limits.

- **Interpretability:** Snippets are readable and auditable.

<div align="center">
<img src="assets/output-example.png" width=800 alt="Output example: it contains a short python snippet as well as its output which is used in the reasoning process." /><br></div>
<em>Figure 2: Output example where python code is generated, evaluated and the answer is inserted into the trace and used for context.</em>

## Training with GRPO

We fine-tune the model using **GRPO**, a reward-based optimization that balances:

- **Accuracy Reward:** +1 for correct answers.

- **Using code snippets:** +1 for generating code snippets, weighted 10:1 vs. the accuracy reward.

- **Length reduction:** shorter lengths are encouraged by limiting the GRPO completion candidates to 5k tokens.

- **Temperature Scheduling:** We implemented linear temperature scheduling (T=1.2 → T=0.7) to balance exploration and stability during training. This approach aims to enhance experimentation during the initial training phases, subsequently reducing the temperature as we refine our proficiency in the skill.

- **In-context Learning**: we include 4 solved examples where the trace contains agent calls and executor outputs, so the model learns the syntax and the call/response pattern.

- **Dataset**: we used [OpenMathReasoning](https://huggingface.co/datasets/nvidia/OpenMathReasoning) dataset, the tool-usage subset. Note that GRPO only uses the <u>problem</u>, not the solution in the data. Choosing this dataset ensures problems benefit form tool use.

## Evaluation

We benchmarked DeepMath against baselines on four datasets. Metrics include:

- **majority@16** (robustness across samples).

- **Mean output length** (brevity).

- We compare a baseline configuration ([Qwen3-4B-Thinking-2507](https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507), no agenting) with our DeepMath model. As ablation, we evaluate the agentic framework we developed running with the untrained Qwen3 model, denoted by **+Agent**. Additionally, we examine whether the GRPO training (for agentic use) improves non-agentic inference, denoted by **+GRPO**. Thus the two ablations are independent, not additive.

- We observe the agentic inference reduces output lengths, with mixed accuracy results. The DeepMath model is both GRPO-trained and run in agentic mode, and shows the highest accuracy with shortened traces. We conclude **both GRPO training and agentic inference are needed** for best results.

**Key Insight:** DeepMath reduces output length by up to **66%** while improving accuracy on challenging datasets.

## Why It Matters

- **Accuracy:** Offloading computation reduces arithmetic errors.

- **Efficiency:** Shorter outputs mean faster inference and easier interpretability.

- **Safety:** Sandbox execution mitigates risks of running arbitrary code.

## Conclusion

DeepMath demonstrates a practical and lightweight way to combine a small executor with an LLM and to train the model to prefer short, computation-driven traces. Offloading deterministic computation reduces arithmetic and numerical errors and shortens traces, and GRPO fine-tuning further encourages concise, correct answers. The result is a more accurate and more interpretable math-solving agent without requiring a massive model or heavyweight external tools.

## Try It Yourself

Check out the GitHub repo and share your feedback! Contributions welcome. 🚀

## Installation

DeepMath uses [uv](https://github.com/astral-sh/uv):

1. **Install uv** (if not already installed):

```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```

2. **Clone the repository**:

```bash
git clone https://github.com/intellabs/DeepMath.git
cd DeepMath
```

3. **Install dependencies**:

```bash
uv pip install -r requirements.txt
uv pip install -e .
```

## Usage

The `example.sh` script provides complete examples for training, inference, and evaluation. Below are the main workflows:

### Training

Training involves two components running in parallel:

1. **Start the vLLM server** (for generating completions during GRPO training):

```bash
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5 \
    python -m deep_math.vllm_serve_agent \
    --model Qwen/Qwen3-4B-Thinking-2507 \
    --tensor_parallel_size 1 \
    --data_parallel_size 6 \
    --config configs/vllm_agent_server_4B.yaml \
    --max_steps 50 \
    --max_agent_output 5000
```

2. **Run the GRPO trainer**:

```bash
CUDA_VISIBLE_DEVICES=6,7 \
    accelerate launch --config_file configs/zero2.yaml \
    --num_processes 2 \
    training.py
```

### Inference

Run inference with different configurations. Examples use the MATH500 dataset:

**Baseline (no agent, no fine-tuning)**:

```bash
python inference.py \
    +model.use_vllm=true \
    model.generation.max_new_tokens=20000 \
    model.generation.do_sample=true \
    model.generation.temperature=0.6 \
    model.model_name_or_path=Qwen/Qwen3-4B-Thinking-2507 \
    hf_tag=HuggingFaceH4/MATH-500 \
    generated_file=baseline-MATH500.jsonl
```

**With agent** (adds Python code execution):

```bash
python inference.py \
    +model.use_vllm=true \
    +model.math_agent=true \
    +model.examples=deep_math/fewshot.txt \
    model.generation.max_new_tokens=3000
    +model.max_agent_output=20000 \
    +model.max_steps=50 \
    model.model_name_or_path=Qwen/Qwen3-4B-Thinking-2507 \
    hf_tag=HuggingFaceH4/MATH-500 \
    generated_file=agent-MATH500.jsonl
```

**With fine-tuned LoRA adapter**:

```bash
python inference.py \
    +model.use_vllm=true \
    +model.math_agent=true \
    model.lora_path=/PATH/TO/CHECKPOINT \
    +model.vllm_params.max_lora_rank=32 \
    model.generation.max_new_tokens=3000
    +model.max_agent_output=20000 \
    +model.max_steps=50 \
    model.model_name_or_path=Qwen/Qwen3-4B-Thinking-2507 \
    hf_tag=HuggingFaceH4/MATH-500 \
    generated_file=trained-agent-MATH500.jsonl
```

### Evaluation

Compare multiple inference runs:

```bash
python evaluation.py \
    generated_file=baseline-MATH500.jsonl,agent-MATH500.jsonl,trained-agent-MATH500.jsonl
```

See `example.sh` for complete command examples with all parameters.

## Citation

If you use DeepMath in your research, please cite:

```bibtex
@software{deepmath2025,
  author = {Fleischer, Daniel and Berchansky, Moshe and Wasserblat, Moshe},
  title = {DeepMath: A Lightweight Math Reasoning Agent for LLMs},
  year = {2025},
  publisher = {Intel AI Labs},
  url = {https://github.com/IntelLabs/DeepMath}
}
```

## Limitations & Future Work

- **Scope**: we focused on a small model and on mathematical reasoning.

- **Generalization**: evaluated on contest-style math; results may not transfer to open-ended mathematical creativity or formal proofs.

- Executing generated code is inherently risky. DeepMath uses strict sandboxing and resource limits, but any deployment should carefully manage attack surfaces and enforce rate limits.

DeepMath

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README