Public record
Software health reportschema 0.34.0 · metrics 2.10.0 · 2026-09-20 02:02 UTC

Lightning-AI / torchmetrics

Machine learning metrics for distributed, scalable PyTorch applications.

PythonApache-2.0★ 2,473 stars⑂ 526 forkssince Dec 2020View on GitHub ↗
KindCommand-line toolLibraryhow this is determined
#python#data-science#machine-learning#pytorch#deep-learning#metrics#analyses

Lightning-AI/torchmetrics holds a health index of 90 out of 100, placing it in the Excellent band. It scores highest on Community & Adoption (92/100) and lowest on Security (40/100). It was last updated 5 days ago. 2 contributors account for most of its recent work.

90
overall / 100
Excellent

Software health index

Metrics are grouped into weighted categories on one standardized 1–100 scale. Overall starts as their weighted mean, calibrated against the distribution of the public record so bands carry percentile meaning; when public evidence triggers the High-Risk Jurisdiction Policy, the rating is adjusted and receives an At Risk ceiling of 34.

90
Exceptional93-100The record's top tier (≈ top 5%); essentially all checked criteria met
Excellent80-92Strong across the board; minor gaps
Good65-79Healthy; gaps are limited and manageable
Moderate50-64Acceptable with notable gaps; review recommended
Weak35-49Material weaknesses across several areas
At Risk20-34Significant weaknesses; adoption warrants caution
Critical1-19Severe problems (abandoned, single-maintainer, no hygiene)
VitalityCommunity &AdoptionSustainability &GovernanceEngineeringQualitySecurityAI Readiness

Score profile

Each axis is a category. The shape matters more than the average — a healthy subject fills the whole shape, while a spike-and-crater profile means strength in one dimension is masking risk in another.

The weighted overall 76 is calibrated to 90 on the published index scale (record calibration 2026-08-02).

Ownership

⚡️ Lightning AI Organization
6,868 followers30 public repossince Dec 2019

This repository is backed by an organization — shared, accountable stewardship that can outlive any single maintainer.

Package ecosystems

RegistryPackageVersionDownloads / moVersionsLast publishTags
PyPItorchmetrics1.9.011,497,31173194 days agodeep-learningmachine-learningpytorchmetricsai

Metrics by category

Vitality

Is the project alive — is code being written and are releases shipping?

75Good · 21% of overall

Development activity

73Good
How it's scored
36/36Push recency — last push 5 days ago
13.8/36Commit cadence — 20/52 weeks with commits
16/18Commit volume — 60 commits in the last year
0/10OpenSSF Scorecard: Maintained — no data
Inputs used
commits_last_year60
human_commit_share0.59
days_since_last_push5
active_weeks_last_year20

Release discipline

78Good
How it's scored
27/27Ships releases — 65 releases published
16.2/36Release recency — latest release 194 days ago
27/27Release cadence — a release every ~40 days
0/10OpenSSF Scorecard: Signed-Releases — no data
Inputs used
releases_count65
latest_release_tagv1.9.0
releases_from_tagsno
days_since_latest_release194
mean_days_between_releases40

Community & Adoption

Does the project have users, downloads, attention, and a welcoming setup for contributors?

92Excellent · 17% of overall

Popularity & adoption

86Excellent
How it's scored
55/60Stars — 2,473 stars
22.7/25Forks — 526 forks
7.8/15Watchers — 26 watchers
Inputs used
forks526
stars2,473
watchers26
growth_stateunverified
growth_factor_pct100
growth_unverified_reasonno_history

Community health

92Excellent
How it's scored
22.5/22.5README
22.5/22.5License — recognized license (Apache-2.0)
18/18CONTRIBUTING guide
13.5/13.5Code of conduct
0/7.2Issue template
6.3/6.3PR template
Inputs used
has_readmeyes
has_licenseyes
readme_badges10
has_contributingyes
has_issue_templateno
has_code_of_conductyes
readme_badge_servicesbadge.fury.io, codecov.io, dev.azure.com, github.com, readthedocs.org, shields.io
has_pull_request_templateyes

Ecosystem adoption (downloads)

100Exceptional
How it's scored
80/80Monthly downloads — 11,497,311 downloads/month across pypi
0/20Registry dependents — not reported by this ecosystem
Inputs used
packagestorchmetrics
dependents—
ecosystemspypi
total_downloads—
monthly_downloads11,497,311
unverified_packages_excluded—
Excluded from scoring (no data or not applicable): Registry dependents. Remaining weights renormalized.

Sustainability & Governance

Will the project survive its people — bus factor, responsiveness, who backs it, and package upkeep?

75Good · 23% of overall

Maintainer resilience (bus factor)

57Moderate
How it's scored
25.2/54Bus factor — 2 contributor(s) cover half of all commits
12.9/22.5Commit distribution — top contributor authored 43% of commits
13.5/13.5Contributor breadth — 97 contributors
0/10OpenSSF Scorecard: Contributors — no data
Inputs used
bus_factor2
contributors_sampled97
top_contributor_share0.426

Issue & PR responsiveness

81Excellent
How it's scored
37.7/42Issue resolution — 90% of issues closed
27/30PR acceptance — 1,924/2,135 decided PRs merged
4.3/13Newcomer PR acceptance — 5/15 first-time contributors' PRs merged in 30d
0/15OpenSSF Scorecard: Code-Review — no data
Inputs used
merged_prs1,924
open_issues107
closed_issues930
prs_merged_7d0
prs_decided_7d0
prs_merged_30d9
prs_decided_30d20
issue_closed_ratio0.897
closed_unmerged_prs211
first_time_authors_30d6
first_time_prs_merged_30d5
first_time_prs_decided_30d15

Ownership & stewardship

78Good
How it's scored
30/30Ownership backing — organization-owned
0/20Verified domain
25/25Owner reach — 6,868 followers of Lightning-AI
22.9/25Track record — 30 public repos, account ~6 yr old
Inputs used
followers6,868
owner_typeOrganization
is_verifiedno
owner_loginLightning-AI
public_repos30
account_age_days2,484

Package maintenance

91Excellent
How it's scored
25/25Published & resolvable — 1 package(s) on pypi
26/35Publish recency — latest publish 194 days ago
20/20Version history — 73 published versions
20/20Not deprecated — active, not deprecated or yanked
Inputs used
packagestorchmetrics
ecosystemspypi
any_deprecatedno
min_days_since_publish194

Engineering Quality

Are baseline engineering and documentation practices in place?

91Excellent · 19% of overall

Engineering practices

92Excellent
How it's scored
24/24CI workflows — 12 workflow(s)
24/24Tests present
16/16Linter config — pyproject.toml ([tool.ruff])
9.6/9.6Pre-commit hooks
0/6.4.editorconfig
0/20OpenSSF Scorecard: CI-Tests — no data
Inputs used
has_ciyes
has_testsyes
has_editorconfigno
has_linter_configyes
has_precommit_configyes

Documentation

90Excellent
How it's scored
30/30README
25/25Documentation directory
15/15Documentation / homepage site — https://lightning.ai/docs/torchmetrics/
10/10Repository description
10/10Topics — 7 topics
0/10Wiki
Inputs used
topicspython, data-science, machine-learning, pytorch, deep-learning, metrics, analyses
has_wikino
homepagehttps://lightning.ai/docs/torchmetrics/
docs_sitehttps://lightning.ai/docs/torchmetrics/
has_readmeyes
has_docs_diryes
has_descriptionyes

Security

Are visible security and supply-chain practices strong, without unresolved high-risk jurisdiction exposure?

40Weak · 16% of overall

Security posture

25At Risk
How it's scored
0/30Security policy (SECURITY.md)
25/25Dependabot config
0/25Dependency lockfiles
0/20CodeQL workflow
Inputs used
sourcefile_signals
lockfiles—
manifestspyproject.toml, requirements.txt, setup.py
has_codeql_workflowno
has_security_policyno
has_dependabot_configyes

Dependency advisories

100Exceptional
How it's scored
35/35Direct dependencies free of known advisories — no direct dependency carries a known advisory
25/25Indirect dependencies free of known advisories — no indirect dependency carries a known advisory
0/40No advisories left outstanding — no advisory carries a publication date
Inputs used
sourceosv
advisories0
affected_packages0
assessed_packages32
unassessed_packages0
affected_by_severitynone
direct_affected_packages0
Excluded from scoring (no data or not applicable): No advisories left outstanding. Remaining weights renormalized. Matched the pypi:torchmetrics@1.9.0 runtime dependency closure — what installing the published package pulls in — 32 packages. Reachability is not analyzed.

AI Readiness

How well is the repo equipped to be developed and maintained with AI coding agents? Carries a deliberately small weight (4%): agent tooling is a real maintenance signal, but a repository with none can still reach 100/100.

84Excellent · 4% of overall

Agent context & guidance

55Moderate
How it's scored
0/45Agent instructions — no CLAUDE.md / AGENTS.md / editor rules
15/15Machine-readable docs (llms.txt) — llms.txt served by the project website (https://lightning.ai/llms.txt)
40/40Legible commit history — 55 of 59 human commits state their intent (structured subject or explanatory body)
Inputs used
has_llms_txtyes
llms_txt_urlhttps://lightning.ai/llms.txt
legible_history_share0.932
agent_instruction_files—
agent_instruction_max_bytes—

Verify loop (build / test / typecheck)

100Exceptional
How it's scored
18/18One-command bootstrap — Makefile, docs/Makefile
22/22Automated tests
11/11Lint / format config — pyproject.toml ([tool.ruff])
11/11Static type checking — src/torchmetrics/py.typed
10/10Reproducible environment — devcontainer, Dockerfile
10/10Demonstrated agent practice — 6 of the last 100 commits agent-authored or agent-credited
8/8Automated maintenance — 41 of the last 100 commits are automated dependency updates
0/10OpenSSF Scorecard: Pinned-Dependencies — no data
Inputs used
has_nixno
has_testsyes
lockfiles—
has_dockerfileyes
typed_languageno
bootstrap_filesMakefile, docs/Makefile
has_devcontaineryes
has_linter_configyes
typecheck_configssrc/torchmetrics/py.typed
agent_commit_share0.06
toolchain_manifests—
dependency_bot_commit_share0.41

Code legibility for models

82Excellent
How it's scored
27/45Type-checkable code — Python with type-check config (src/torchmetrics/py.typed)
55/55Manageable file sizes — 0/563 source files over 60KB
Inputs used
primary_languagePython
largest_source_bytes56,859
source_files_sampled563
oversized_source_files0

Machine-readable interfaces

100Exceptional
How it's scored
0/40API schema (OpenAPI/GraphQL/proto) — not applicable to this kind of software
0/20MCP server — not applicable to this kind of software
40/40Runnable examples — examples
Inputs used
example_dirsexamples
has_mcp_signalno
api_schema_files—
interfaces_expected_of—
Excluded from scoring (no data or not applicable): API schema (OpenAPI/GraphQL/proto), MCP server. Remaining weights renormalized.

Key facts

2,473GitHub stars
97contributors
60commits, last 12 months
5days since last push
65releases
2bus factor
107open issues
PyPIpackage ecosystems

Data collection warnings

  • Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token
  • OpenSSF Scorecard did not return a usable result (killed by SIGKILL — most likely the container's memory limit; 2026/09/20 02:01:53 Warning: PATs stored in env variables GITHUB_AUTH_TOKEN and GITHUB_TOKEN differ. Scorecard will use the former.); skipping Scorecard checks

More detail

Star and fork history 0 ★ / 526 ⇿
0Stars
526Forks
65Releases

When each star and fork was added, collected from GitHub and bucketed by day. Cumulative growth sits directly above the daily additions it is made of, so the two read against each other: steady organic accretion looks nothing like an abrupt, short-lived burst. Where that difference is measurable, it is reported as growth authenticity.

010020030040050060052082021-032023-122026-09
Major 1Minor 19Patch 43

Each point covers 6 days.

All dependencies 106

Full resolved dependency set from the GitHub dependency graph: 0 direct and 106 indirect (transitive) packages. The transitive closure is complete when the repository commits a lockfile.

RegistryPackageVersionRelation
PyPIaeon—indirect
PyPIbert-score0.3.13indirect
PyPIcachier4.2.0indirect
PyPIcloudpickle—indirect
PyPIcodecov2.1.13indirect
PyPIcoverage7.10.7indirect
PyPIdists-pytorch0.1indirect
PyPIdocutils0.19indirect
PyPIdython0.7.9indirect
PyPIeinops—indirect
PyPIfairlearn—indirect
PyPIfast-bss-eval—indirect
PyPIfaster-coco-eval—indirect
PyPIfire—indirect
PyPIgammatone—indirect
PyPIhuggingface-hub—indirect
PyPIipadic—indirect
PyPIipython—indirect
PyPIjiwer—indirect
PyPIkornia—indirect
PyPIlai-sphinx-theme—indirect
PyPIlibrosa—indirect
PyPIlightning—indirect
PyPIlightning-utilities—indirect
PyPIlightning-utilities0.15.3indirect
PyPIlpips—indirect
PyPImatplotlib—indirect
PyPImecab-ko—indirect
PyPImecab-ko-dic—indirect
PyPImecab-python3—indirect
PyPImir-eval—indirect
PyPImonai1.4.0indirect
PyPImypy1.18.2indirect
PyPImyst-parser1.0.0indirect
PyPInetcal—indirect
PyPInltk—indirect
PyPInumpy—indirect
PyPIonnxruntime—indirect
PyPIpackaging—indirect
PyPIpandas—indirect
PyPIpandoc2.4indirect
PyPIpermetrics2.0.0indirect
PyPIpesq—indirect
PyPIphmdoctest1.4.0indirect
PyPIpiq—indirect
PyPIproperscoring0.1indirect
PyPIpsutil—indirect
PyPIpy-tree—indirect
PyPIpycocotools—indirect
PyPIpydantic—indirect
PyPIpygithub—indirect
PyPIpystoi—indirect
PyPIpytdc—indirect
PyPIpytest—indirect
PyPIpytest8.4.2indirect
PyPIpytest-cov7.0.0indirect
PyPIpytest-doctestplus—indirect
PyPIpytest-doctestplus1.4.0indirect
PyPIpytest-rerunfailures—indirect
PyPIpytest-rerunfailures16.0.1indirect
PyPIpytest-timeout2.4.0indirect
PyPIpytest-xdist3.8.0indirect
PyPIpytorch-lightning—indirect
PyPIpytorch-msssim1.0.0indirect
PyPIregex—indirect
PyPIrequests—indirect
PyPIrouge-score—indirect
PyPIsacrebleu—indirect
PyPIsafetensors—indirect
PyPIscienceplots—indirect
PyPIscikit-image—indirect
PyPIscikit-learn—indirect
PyPIscipy—indirect
PyPIsentencepiece—indirect
PyPIsetuptools—indirect
PyPIsewar—indirect
PyPIsphinx5.3.0indirect
PyPIsphinx-autobuild2024.10.3indirect
PyPIsphinx-autodoc-typehints1.23.0indirect
PyPIsphinx-copybutton0.5.2indirect
PyPIsphinx-gallery0.19.0indirect
PyPIsphinx-paramlinks0.6.0indirect
PyPIsphinx-togglebutton0.3.2indirect
PyPIsphinxcontrib-fulltoc—indirect
PyPIsphinxcontrib-mockautodoc—indirect
PyPIsrmrpy—indirect
PyPIstatsmodels—indirect
PyPItimm—indirect
PyPItorch—indirect
PyPItorch2.10.0indirect
PyPItorch-complex—indirect
PyPItorch-fidelity—indirect
PyPItorch-linear-assignment—indirect
PyPItorchaudio—indirect
PyPItorchvision—indirect
PyPItqdm—indirect
PyPItransformers—indirect
PyPItypes-emoji—indirect
PyPItypes-protobuf—indirect
PyPItypes-pyyaml—indirect
PyPItypes-requests—indirect
PyPItypes-setuptools—indirect
PyPItypes-six—indirect
PyPItypes-tabulate—indirect
PyPIvmaf-torch—indirect
PyPIwget—indirect
Dependency advisories 0

Installing pypi:torchmetrics@1.9.0 pulls in 32 packages, direct and transitive: 0 carry known advisories, of which 0 are direct dependencies.

No known advisories affect the assessed dependencies.

An advisory means the version recorded in the dependency graph falls inside an advisory’s affected range. Reachability is not analysed, and the graph includes development and test pins — a finding may concern tooling rather than shipped software.

Raw JSON report machine-readable

Feedback

Spotted something off in this report, or have thoughts to share? Wrong measurements, missed tooling, ideas, questions — anything is welcome. Every message is read and gets a response.

The message is kept through sign-in.

Related records

Inspected repositories sharing catalogue tags or ecosystems with Lightning-AI/torchmetrics. Read them side by side →

100wandb/wandbThe AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.Exceptional99ultralytics/ultralyticsUltralytics YOLO27, YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object trackingExceptional99ray-project/rayRay is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.Exceptional98autogluon/autogluonFast and Accurate ML in 3 Lines of CodeExceptional98zenml-io/zenmlZenML 🙏: One AI Platform from Pipelines to Agents. https://zenml.io.Exceptional98pytorch/igniteHigh-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.Exceptional

Scores are signals, not warranties. They reflect publicly visible practices on GitHub — not a code audit, and not a security guarantee.

Missing data is excluded and weights renormalized, never scored as zero. Methodology is versioned and open: metrics v2.10.0, schema v0.34.0 — full methodology · metrics wiki.

How one result sits in the wider record: aggregate statistics — PyPI.