# QwenLM/Qwen2.5-Math

A series of math-specific large language models of our Qwen2 series.

Repository: https://github.com/QwenLM/Qwen2.5-Math
Canonical: https://ross.abutalabs.com/products/qwen25-math
Homepage: https://qwenlm.github.io/blog/qwen2-math/
Language: Python
License Family: other
Topics: large-language-models, mathematics, qwen2
Last push: 2025-01-11T17:30:43+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 1, release rhythm 35, longevity 53
- inputs: {"age_days": 755, "days_push": 599, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1087, forks 160 (observed 2026-08-28T04:03:32.337673+00:00)

## What it is
Qwen2.5-Math is a series of math-specialized large language models (1.5B/7B/72B base and instruct variants plus a 72B reward model) built on the Qwen2 architecture. The repository provides inference code, quickstart examples, and evaluation tooling for solving English and Chinese math problems via chain-of-thought and tool-integrated reasoning.

## Use cases
- solve math word problems with an llm
- run a math-specialized language model locally
- evaluate llm performance on gsm8k and math benchmarks
- use tool-integrated reasoning for math with a python interpreter
- fine-tune or build on math-focused open weights
- score math solutions with a reward model

## When to choose
- you need strong math problem-solving in English or Chinese
- you want open-weight models with CoT and tool-integrated reasoning
- you need a reward model for math RL pipelines

## When to avoid
- you need a general-purpose assistant for non-math tasks
- you lack GPU resources for the larger checkpoints
- you need a permissive license - the repo lists no license

## Facets
- artifact type: library
- maturity: active
- function: llm-inference, machine-learning, nlp
- domain: large-language-models, mathematics
- platform: python
- tags: math-reasoning, chain-of-thought, tool-integrated-reasoning, qwen, open-weights, reward-model, natural-language-processing, gpu, linux

## Member repositories
- QwenLM/Qwen2.5-Math (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:32.337673+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:49:25.478156+00:00, confidence not recorded.
  - readme: https://github.com/QwenLM/Qwen2.5-Math (fetched 2026-08-28T04:03:32.337673+00:00, sha 4aeefe15962f)
  - homepage: https://qwenlm.github.io/blog/qwen2-math/ (fetched 2026-08-29T12:52:09.736291+00:00, sha ce0201550704)
- Data as of 2026-08-30T08:39:29.467469+00:00.