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.

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

Visit source ↗huggingface.co

Announced: 04 Dec 2025 00:00 UTC
Detected: 03 Aug 2026 21:12 UTC (+242d after)

Other stories

Hugging Face · 28 AugThe Open ASR Leaderboard Adds Its First Global South LanguageCapability change

Google Research · 27 AugPlanetary prediction engine: Automating global models via Earth AIResearch paper

OpenAI · 27 AugBetter answers, broader thinking: What students gain from ChatGPT and critical-thinking trainingResearch paper

LG AI Research · 27 Aug

[ACL 2026] ReSQL: A Self-Improving Text-to-SQL Framework through Retrieval-Augmented Error Reasoning

Research paper

xAI · 29 AugGrok Bot now works with XCapability change

GitHub · 28 AugGitHub Copilot in Visual Studio — August update