# openai/grade-school-math

Repository: https://github.com/openai/grade-school-math
Canonical: https://ross.abutalabs.com/products/grade-school-math
Language: Python
License Family: other
Archived: true
Last push: 2024-01-21T10:54:11+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1778, "days_push": 955, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1451, forks 200 (observed 2026-08-28T04:04:45.953799+00:00)

## What it is
GSM8K is a dataset of 8.5K linguistically diverse grade school math word problems (7.5K train, 1K test) released by OpenAI to diagnose multi-step mathematical reasoning in language models. The repository also includes helper code for calculator-annotated sampling and solution extraction.

## Use cases
- benchmark llm math reasoning
- evaluate chain-of-thought on word problems
- finetune models on multi-step arithmetic
- test calculator-annotated sampling
- research multi-step reasoning failures

## When to choose
- you need a standard math reasoning benchmark for language models
- you want human-written grade school word problems with verifiable numeric answers
- you are reproducing or extending the GSM8K paper

## When to avoid
- you need actively maintained tooling or support
- you need advanced mathematics beyond grade school arithmetic
- you need a general-purpose math solver rather than evaluation data

## Facets
- artifact type: dataset
- maturity: abandoned
- function: machine-learning, benchmarking, math, data-science
- domain: large-language-models, machine-learning, mathematics
- platform: python
- tags: gsm8k, math-word-problems, reasoning-benchmark, llm-evaluation, archived, natural-language-processing

## Member repositories
- openai/grade-school-math (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:45.953799+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 2026-08-30T04:35:52.174203+00:00, confidence not recorded.
  - readme: https://github.com/openai/grade-school-math (fetched 2026-08-28T04:04:45.953799+00:00, sha 0053c3db8996)
- Data as of 2026-08-30T08:39:29.467469+00:00.