# Qwen2.5-Math series open-sourced with reward model

> Qwen upgraded their math-focused LLM series to Qwen2.5-Math, releasing base models (1.5B/7B/72B), instruction-tuned variants, and a mathematical reward model. The models primarily support English and Chinese math problems via chain-of-thought and tool-integrated reasoning.

| | |
|---|---|
| **Tool** | Qwen |
| **Version** | — |
| **Kind** | announcement |
| **Published** | 2024-09-18 |
| **Observed** | 2026-08-11 |
| **Significance** | 4/5 |
| **Breaking** | no |
| **Categories** | new-model, feature |

## What changed


- Base models: Qwen2.5-Math-1.5B/7B/72B
- Instruction-tuned models: Qwen2.5-Math-1.5B/7B/72B-Instruct
- Mathematical reward model also open-sourced
- Supports English and Chinese math problems through CoT and TIR
- Not recommended for non-math tasks


## Sources

- [blog_rss](https://qwenlm.github.io/blog/qwen2.5-math/) — retrieved 2026-08-11


## Community

_No curated reactions recorded. Facts and community takes are kept in separate layers
and never blended._

---
Canonical: https://changelogs.info/qwen/qwen2-5-math-series-open-sourced-with-reward-model
Entity: https://changelogs.info/qwen
Event ID: `evt_2024-09-18_qwen_qwen2-5-math-the-world-s-leading-open-sourced-mathematical-llms`
Licence: event synthesis © changelogs.info, CC BY 4.0. Linked sources belong to their vendors.