# Hugging Face — DeepMath: A lightweight math reasoning Agent with smolagents

- Company: Hugging Face (huggingface.co)
- Announced: 2025-12-04T00:00:00+00:00
- Category: research-paper
- Subject: Platform
- Models affected: Qwen3-4B Thinking, deepmath-v1
- Source: https://huggingface.co/blog/intel-deepmath
- Record: https://forck.live/items/1575-deepmath-a-lightweight-math-reasoning-agent-with-smolagents

Intel AI Software Group introduces DeepMath, a math reasoning agent based on Qwen3-4B Thinking fine-tuned with GRPO. The model emits Python snippets for intermediate steps, runs them in a sandbox, and uses the results in its reasoning. It reduces output lengths by up to 66% and improves accuracy on math datasets.

## Evidence

Verbatim from https://huggingface.co/blog/intel-deepmath:

> DeepMath is an aligned math reasoning agent built on Qwen3-4B Thinking and fine-tuned with GRPO (Group Relative Policy Optimization).

---

Record: https://forck.live/items/1575-deepmath-a-lightweight-math-reasoning-agent-with-smolagents
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md