# google-deepmind/alphatensor

Repository: https://github.com/google-deepmind/alphatensor
Canonical: https://ross.abutalabs.com/products/alphatensor
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-04-22T20:57:33+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1441, "days_push": 863, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2853, forks 257 (observed 2026-08-28T04:07:25.734674+00:00)

## What it is
Code and data accompanying DeepMind's AlphaTensor publication on discovering faster matrix multiplication algorithms with reinforcement learning. It includes discovered algorithms as tensor factorizations, GPU benchmarking scripts, nonequivalence verification data, and recombination code.

## Use cases
- explore discovered matrix multiplication algorithms
- benchmark matrix multiplication speed on NVIDIA GPUs
- verify nonequivalence of matrix multiplication factorizations
- decompose larger matrix multiplication tensors from smaller factorizations
- study reinforcement learning for algorithm discovery

## When to choose
- you want to inspect or use AlphaTensor's discovered matrix multiplication algorithms
- you need to benchmark matrix multiplication algorithms on a V100 GPU
- you are reproducing or extending the AlphaTensor research

## When to avoid
- you need a production matrix multiplication library for your application
- you want the full AlphaTensor training pipeline, which is not included
- you need a maintained general-purpose linear algebra tool

## Facets
- artifact type: dataset
- maturity: maintenance
- function: machine-learning, reinforcement-learning, benchmarking, math
- domain: machine-learning
- platform: python
- tags: matrix-multiplication, alphazero, research-code, tensor-decomposition, deepmind, scientific-publication, algorithms, research, linux, gpu

## Member repositories
- google-deepmind/alphatensor (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:25.734674+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 2026-08-30T07:36:46.852556+00:00, confidence not recorded.
  - readme: https://github.com/google-deepmind/alphatensor (fetched 2026-08-28T04:07:25.734674+00:00, sha 153ecc694ef3)
- Data as of 2026-08-30T08:39:29.467469+00:00.