Software health reportschema 0.31.0 · metrics 2.10.0 · 2026-08-12 20:32 UTC

NVIDIA / Megatron-LM

Ongoing research training transformer models at scale

PythonCustom license★ 17,403 stars⑂ 4,357 forkssince Mar 2019View on GitHub ↗

KindLibraryCommand-line toolhow this is determined

#large-language-models #model-para #transformers

NVIDIA/Megatron-LM holds a health index of 98 out of 100, placing it in the Exceptional band. It scores highest on Vitality (100/100) and lowest on Security (64/100). It was last updated today. 6 contributors account for most of its recent work.

overall / 100

Exceptional

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.

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)

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 87 is calibrated to 98 on the published index scale (record calibration 2026-08-02).

Ownership

NVIDIA CorporationOrganization

28,894 followers785 public repossince May 2012

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

Package ecosystems

Registry	Package	Version	Downloads / mo	Versions	Last publish	Tags
PyPI	megatron-core	0.18.2	234,379	393	22 days ago	nlp nlu deep gpu language learning machine nvidia pytorch torch transformer

Metrics by category

Vitality

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

100Exceptional · 21% of overall

Development activity

100Exceptional

How it's scored

36/36	Push recency — last push 0 days ago
36/36	Commit cadence — 52/52 weeks with commits
18/18	Commit volume — 2,488 commits in the last year
0/10	OpenSSF Scorecard: Maintained — no data

Inputs used

commits_last_year	2,488
human_commit_share	0.94
days_since_last_push	0
active_weeks_last_year	52

Release discipline

100Exceptional

How it's scored

27/27	Ships releases — 47 releases published
36/36	Release recency — latest release 22 days ago
27/27	Release cadence — a release every ~21.5 days
0/10	OpenSSF Scorecard: Signed-Releases — no data

Inputs used

releases_count	47
latest_release_tag	core_v0.18.2
releases_from_tags	no
days_since_latest_release	22
mean_days_between_releases	21.5

Community & Adoption

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

86Excellent · 17% of overall

Popularity & adoption

97Exceptional

How it's scored

60/60	Stars — 17,403 stars
25/25	Forks — 4,357 forks
12.3/15	Watchers — 165 watchers

Inputs used

forks	4,357
stars	17,403
watchers	165
growth_state	unverified
growth_factor_pct	100
growth_unverified_reason	no_history

Community health

71Good

How it's scored

22.5/22.5	README
16.9/22.5	License — license file present, not a recognized license
18/18	CONTRIBUTING guide
0/13.5	Code of conduct
0/7.2	Issue template
6.3/6.3	PR template

Inputs used

has_readme	yes
has_license	yes
readme_badges	3
has_contributing	yes
has_issue_template	no
has_code_of_conduct	no
readme_badge_services	shields.io
has_pull_request_template	yes

Ecosystem adoption (downloads)

89Excellent

How it's scored

71.6/80	Monthly downloads — 234,379 downloads/month across pypi
0/20	Registry dependents — not reported by this ecosystem

Inputs used

packages	megatron-core
dependents	—
ecosystems	pypi
total_downloads	—
monthly_downloads	234,379
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?

88Excellent · 23% of overall

Maintainer resilience (bus factor)

88Excellent

How it's scored

48.6/54	Bus factor — 6 contributor(s) cover half of all commits
17.4/22.5	Commit distribution — top contributor authored 22% of commits
13.5/13.5	Contributor breadth — 99 contributors
0/10	OpenSSF Scorecard: Contributors — no data

Inputs used

bus_factor	6
contributors_sampled	99
top_contributor_share	0.225

Issue & PR responsiveness

68Good

How it's scored

32.4/42	Issue resolution — 77% of issues closed
19.2/30	PR acceptance — 2,473/3,860 decided PRs merged
6.5/13	Newcomer PR acceptance — 1/2 first-time contributors' PRs merged in 30d
0/15	OpenSSF Scorecard: Code-Review — no data

Inputs used

merged_prs	2,473
open_issues	382
closed_issues	1,287
prs_merged_7d	41
prs_decided_7d	57
prs_merged_30d	43
prs_decided_30d	59
issue_closed_ratio	0.771
closed_unmerged_prs	1,387
first_time_authors_30d	2
first_time_prs_merged_30d	1
first_time_prs_decided_30d	2

Ownership & stewardship

100Exceptional

How it's scored

30/30	Ownership backing — organization-owned
0/20	Verified domain — verified-domain status not read for this organization
25/25	Owner reach — 28,894 followers of NVIDIA
25/25	Track record — 785 public repos, account ~14 yr old

Inputs used

followers	28,894
owner_type	Organization
is_verified	—
owner_login	NVIDIA
public_repos	785
account_age_days	5,206

Excluded from scoring (no data or not applicable): Verified domain. Remaining weights renormalized.

Package maintenance

100Exceptional

How it's scored

25/25	Published & resolvable — 1 package(s) on pypi
35/35	Publish recency — latest publish 22 days ago
20/20	Version history — 393 published versions
20/20	Not deprecated — active, not deprecated or yanked

Inputs used

packages	megatron-core
ecosystems	pypi
any_deprecated	no
min_days_since_publish	22

Engineering Quality

Are baseline engineering and documentation practices in place?

91Excellent · 19% of overall

Engineering practices

92Excellent

How it's scored

24/24	CI workflows — 33 workflow(s)
24/24	Tests present
16/16	Linter config — .pylintrc
9.6/9.6	Pre-commit hooks
0/6.4	.editorconfig
0/20	OpenSSF Scorecard: CI-Tests — no data

Inputs used

has_ci	yes
has_tests	yes
has_editorconfig	no
has_linter_config	yes
has_precommit_config	yes

Documentation

90Excellent

How it's scored

30/30	README
25/25	Documentation directory
15/15	Documentation / homepage site — https://docs.nvidia.com/megatron-core/developer-guide/latest/get-started/quickstart.html
10/10	Repository description
10/10	Topics — 3 topics
0/10	Wiki

Inputs used

topics	large-language-models, model-para, transformers
has_wiki	no
homepage	https://docs.nvidia.com/megatron-core/developer-guide/latest/get-started/quickstart.html
docs_site	https://docs.nvidia.com/megatron-core/developer-guide/latest/get-started/quickstart.html
has_readme	yes
has_docs_dir	yes
has_description	yes

Security

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

64Moderate · 16% of overall

Security posture

55Moderate

How it's scored

30/30	Security policy (SECURITY.md)
0/25	Dependabot config
25/25	Dependency lockfiles — uv.lock
0/20	CodeQL workflow

Inputs used

source	file_signals
lockfiles	uv.lock
manifests	pyproject.toml, setup.py
has_codeql_workflow	no
has_security_policy	yes
has_dependabot_config	no

Dependency advisories

100Exceptional

How it's scored

35/35	Direct dependencies free of known advisories — no direct dependency carries a known advisory
25/25	Indirect dependencies free of known advisories — no indirect dependency carries a known advisory
0/40	No advisories left outstanding — no advisory carries a publication date

Inputs used

source	osv
advisories	0
affected_packages	0
assessed_packages	3
unassessed_packages	0
affected_by_severity	none
direct_affected_packages	0

Excluded from scoring (no data or not applicable): No advisories left outstanding. Remaining weights renormalized. Matched the pypi:megatron-core@0.18.2 runtime dependency closure — what installing the published package pulls in — 3 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.

80Excellent · 4% of overall

Agent context & guidance

85Excellent

How it's scored

45/45	Agent instructions — .cursorrules, AGENTS.md, CLAUDE.md
0/15	Machine-readable docs (llms.txt)
40/40	Legible commit history — 87 of 94 human commits state their intent (structured subject or explanatory body)

Inputs used

has_llms_txt	no
llms_txt_url	—
legible_history_share	0.926
agent_instruction_files	.cursorrules, AGENTS.md, CLAUDE.md
agent_instruction_max_bytes	3,012

Verify loop (build / test / typecheck)

79Good

How it's scored

18/18	One-command bootstrap — megatron/core/datasets/Makefile
22/22	Automated tests
11/11	Lint / format config — .pylintrc
0/11	Static type checking
10/10	Reproducible environment — Dockerfile, lockfile
10/10	Demonstrated agent practice — 13 of the last 100 commits agent-authored or agent-credited
0/8	Automated maintenance — no automated dependency updates observed
0/10	OpenSSF Scorecard: Pinned-Dependencies — no data

Inputs used

has_nix	no
has_tests	yes
lockfiles	uv.lock
has_dockerfile	yes
typed_language	no
bootstrap_files	megatron/core/datasets/Makefile
has_devcontainer	no
has_linter_config	yes
typecheck_configs	—
agent_commit_share	0.13
toolchain_manifests	—
dependency_bot_commit_share	0

Code legibility for models

53Moderate

How it's scored

0/45	Type-checkable code — Python without a type-check config
52.9/55	Manageable file sizes — 48/1,285 source files over 60KB

Inputs used

primary_language	Python
largest_source_bytes	262,053
source_files_sampled	1,285
oversized_source_files	48

Machine-readable interfaces

100Exceptional

How it's scored

0/40	API schema (OpenAPI/GraphQL/proto) — not applicable to this kind of software
0/20	MCP server — not applicable to this kind of software
40/40	Runnable examples — examples, recipes

Inputs used

example_dirs	examples, recipes
has_mcp_signal	no
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

17,403GitHub stars

99contributors

2,488commits, last 12 months

0days since last push

47releases

6bus factor

382open issues

PyPIpackage ecosystems

Data collection warnings

Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token
First-time contributor figures cover 12 of 31 authors (cap 12)
OpenSSF Scorecard did not return a usable result (2026/08/12 20:31:06 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 ★ / 4,357 ⇿

0Stars

4,357Forks

12Releases

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.

Only the most recent history is shown — this repository exceeds the collection window, so the earliest history is not captured.

Major 0Minor 0Patch 0

Direct dependencies 3

Registry	Package	Version constraint	Manifest
PyPI	torch	`>=2.6.0`	pyproject.toml
PyPI	numpy	`—`	pyproject.toml
PyPI	packaging	`>=24.2`	pyproject.toml

All dependencies 320

Full resolved dependency set from the GitHub dependency graph: 6 direct and 314 indirect (transitive) packages. The transitive closure is complete when the repository commits a lockfile.

Registry	Package	Version	Relation
PyPI	numpy	`—`	direct
PyPI	numpy	`2.5.2`	direct
PyPI	packaging	`—`	direct
PyPI	packaging	`26.3`	direct
PyPI	torch	`—`	direct
PyPI	torch	`2.13.0`	direct
PyPI	absl-py	`2.5.0`	indirect
PyPI	accelerate	`1.14.0`	indirect
PyPI	accessible-pygments	`0.0.5`	indirect
PyPI	aiobotocore	`3.9.0`	indirect
PyPI	aiofiles	`25.1.0`	indirect
PyPI	aiohappyeyeballs	`2.7.1`	indirect
PyPI	aiohttp	`3.14.3`	indirect
PyPI	aiohttp-cors	`0.8.1`	indirect
PyPI	aioitertools	`0.13.0`	indirect
PyPI	aiosignal	`1.4.0`	indirect
PyPI	alabaster	`1.0.0`	indirect
PyPI	aniso8601	`10.0.1`	indirect
PyPI	annotated-doc	`0.0.5`	indirect
PyPI	annotated-types	`0.8.0`	indirect
PyPI	antlr4-python3-runtime	`4.9.3`	indirect
PyPI	anyio	`4.9.0`	indirect
PyPI	apache-tvm-ffi	`0.1.9`	indirect
PyPI	astroid	`3.2.4`	indirect
PyPI	attrs	`26.1.0`	indirect
PyPI	av	`17.0.1`	indirect
PyPI	av	`18.0.0`	indirect
PyPI	babel	`2.18.0`	indirect
PyPI	bcrypt	`5.0.0`	indirect
PyPI	beautifulsoup4	`4.15.0`	indirect
PyPI	black	`26.3.0`	indirect
PyPI	blinker	`1.9.0`	indirect
PyPI	boto3	`—`	indirect
PyPI	botocore	`1.43.56`	indirect
PyPI	braceexpand	`0.1.7`	indirect
PyPI	bracex	`3.0.1`	indirect
PyPI	catalogue	`2.0.10`	indirect
PyPI	causal-conv1d	`—`	indirect
PyPI	causal-conv1d	`1.6.2.post1`	indirect
PyPI	certifi	`2026.7.22`	indirect
PyPI	cffi	`2.1.1`	indirect
PyPI	charset-normalizer	`3.4.9`	indirect
PyPI	click	`8.4.2`	indirect
PyPI	cloudpickle	`3.0.0`	indirect
PyPI	colorama	`0.4.6`	indirect
PyPI	colorful	`0.5.8`	indirect
PyPI	contextlib2	`21.6.0`	indirect
PyPI	coverage	`7.15.4`	indirect
PyPI	cryptography	`50.0.0`	indirect
PyPI	cuda-bindings	`13.3.1`	indirect
PyPI	cuda-core	`1.0.1`	indirect
PyPI	cuda-pathfinder	`1.6.0`	indirect
PyPI	cuda-python	`13.3.1`	indirect
PyPI	cuda-tile	`1.5.0`	indirect
PyPI	cython	`3.2.9`	indirect
PyPI	datasets	`—`	indirect
PyPI	datasets	`4.8.5`	indirect
PyPI	datasets	`5.0.1`	indirect
PyPI	decorator	`5.3.1`	indirect
PyPI	deep-gemm	`2.5.0+local`	indirect
PyPI	defusedxml	`0.7.1`	indirect
PyPI	deprecated	`1.3.1`	indirect
PyPI	deprecation	`2.1.0`	indirect
PyPI	dill	`0.4.1`	indirect
PyPI	diskcache	`—`	indirect
PyPI	distlib	`0.4.3`	indirect
PyPI	distro	`1.9.0`	indirect
PyPI	docker	`7.2.0`	indirect
PyPI	docstring-parser	`0.18.0`	indirect
PyPI	docutils	`0.22.4`	indirect
PyPI	einops	`—`	indirect
PyPI	einops	`0.8.2`	indirect
PyPI	emerging-optimizers	`—`	indirect
PyPI	emerging-optimizers	`0.2.0`	indirect
PyPI	exceptiongroup	`1.3.1`	indirect
PyPI	fabric	`3.2.3`	indirect
PyPI	fast-hadamard-transform	`1.0.4.post1`	indirect
PyPI	fastapi	`0.136.3`	indirect
PyPI	fastapi	`0.141.1`	indirect
PyPI	fiddle	`0.3.0`	indirect
PyPI	filelock	`3.32.2`	indirect
PyPI	filetype	`1.2.0`	indirect
PyPI	fla-core	`0.5.1`	indirect
PyPI	flake8	`7.1.0`	indirect
PyPI	flash-linear-attention	`0.5.1`	indirect
PyPI	flash-mla	`1.0.0+9edee0c`	indirect
PyPI	flashinfer-python	`0.6.11.post3`	indirect
PyPI	flashinfer-python	`0.6.16.post3`	indirect
PyPI	flask	`3.1.3`	indirect
PyPI	flask-restful	`0.3.10`	indirect
PyPI	frozenlist	`1.8.0`	indirect
PyPI	fsspec	`2026.6.0`	indirect
PyPI	google-api-core	`2.34.0`	indirect
PyPI	google-auth	`2.56.3`	indirect
PyPI	google-cloud-language	`—`	indirect
PyPI	googleapis-common-protos	`1.75.1`	indirect
PyPI	graphviz	`0.21`	indirect
PyPI	grpcio	`1.83.0`	indirect
PyPI	grpcio-tools	`1.83.0`	indirect
PyPI	h11	`0.16.0`	indirect
PyPI	h2	`4.4.1`	indirect
PyPI	hatchling	`1.31.0`	indirect
PyPI	hf-xet	`1.6.0`	indirect
PyPI	hpack	`4.2.0`	indirect
PyPI	httpcore	`1.0.9`	indirect
PyPI	httpx	`0.27.2`	indirect
PyPI	httpx-aiohttp	`0.2.0`	indirect
PyPI	huggingface-hub	`1.27.0`	indirect
PyPI	hypercorn	`0.18.0`	indirect
PyPI	hyperframe	`6.1.0`	indirect
PyPI	idna	`3.18`	indirect
PyPI	imagesize	`2.0.0`	indirect
PyPI	importlib-metadata	`8.6.1`	indirect
PyPI	inflect	`—`	indirect
PyPI	iniconfig	`2.3.0`	indirect
PyPI	inquirerpy	`0.3.4`	indirect
PyPI	invoke	`2.2.1`	indirect
PyPI	isort	`5.13.2`	indirect
PyPI	itsdangerous	`2.2.0`	indirect
PyPI	jinja2	`3.1.6`	indirect
PyPI	jiter	`0.16.0`	indirect
PyPI	jmespath	`1.1.0`	indirect
PyPI	joblib	`1.5.3`	indirect
PyPI	jsonschema	`4.26.0`	indirect
PyPI	jsonschema-specifications	`2025.9.1`	indirect
PyPI	lark	`1.3.1`	indirect
PyPI	ledoc-ui	`0.1.0`	indirect
PyPI	leptonai	`0.27.3`	indirect
PyPI	libcst	`1.9.0`	indirect
PyPI	loguru	`0.7.3`	indirect
PyPI	mamba-ssm	`—`	indirect
PyPI	mamba-ssm	`2.3.2.post1`	indirect
PyPI	markdown	`3.10.3`	indirect
PyPI	markdown-it-py	`4.2.0`	indirect
PyPI	markupsafe	`3.0.3`	indirect
PyPI	mccabe	`0.7.0`	indirect
PyPI	mdit-py-plugins	`0.6.1`	indirect
PyPI	mdurl	`0.1.2`	indirect
PyPI	megatron-energon	`7.3.2`	indirect
PyPI	megatron-energon	`7.4.1`	indirect
PyPI	mfusepy	`3.1.1`	indirect
PyPI	ml-dtypes	`0.5.4`	indirect
PyPI	mock	`5.2.0`	indirect
PyPI	mpmath	`1.3.0`	indirect
PyPI	msgpack	`1.2.1`	indirect
PyPI	multi-storage-client	`0.49.0`	indirect
PyPI	multi-storage-client	`0.51.0`	indirect
PyPI	multidict	`6.7.1`	indirect
PyPI	multiprocess	`0.70.19`	indirect
PyPI	mypy-extensions	`1.1.0`	indirect
PyPI	myst-parser	`5.1.0`	indirect
PyPI	nccl4py	`0.3.1`	indirect
PyPI	nemo-run	`0.11.0+e393539`	indirect
PyPI	networkx	`3.6.1`	indirect
PyPI	ninja	`1.13.0`	indirect
PyPI	nltk	`—`	indirect
PyPI	nltk	`3.10.2`	indirect
PyPI	nvdlfw-inspect	`0.2.2`	indirect
PyPI	nvidia-cudnn-frontend	`1.26.0`	indirect
PyPI	nvidia-cutlass-dsl	`4.5.0`	indirect
PyPI	nvidia-cutlass-dsl-libs-base	`4.5.0`	indirect
PyPI	nvidia-cutlass-dsl-libs-cu13	`4.5.0`	indirect
PyPI	nvidia-mathdx	`25.6.0`	indirect
PyPI	nvidia-ml-py	`13.610.43`	indirect
PyPI	nvidia-modelopt	`—`	indirect
PyPI	nvidia-modelopt	`0.45.0`	indirect
PyPI	nvidia-resiliency-ext	`0.6.0`	indirect
PyPI	nvidia-sphinx-theme	`0.0.9.post1`	indirect
PyPI	omegaconf	`—`	indirect
PyPI	omegaconf	`2.3.1`	indirect
PyPI	onnx	`1.22.0`	indirect
PyPI	onnx-ir	`0.2.1`	indirect
PyPI	onnxscript	`0.7.0`	indirect
PyPI	onnxscript	`0.7.1`	indirect
PyPI	openai	`2.53.0`	indirect
PyPI	opencensus	`0.11.4`	indirect
PyPI	opencensus-context	`0.1.3`	indirect
PyPI	opentelemetry-api	`1.33.1`	indirect
PyPI	opentelemetry-exporter-prometheus	`0.54b1`	indirect
PyPI	opentelemetry-proto	`1.44.0`	indirect
PyPI	opentelemetry-sdk	`1.33.1`	indirect
PyPI	opentelemetry-semantic-conventions	`0.54b1`	indirect
PyPI	orjson	`3.11.9`	indirect
PyPI	pandas	`—`	indirect
PyPI	pandas	`3.0.5`	indirect
PyPI	paramiko	`5.0.0`	indirect
PyPI	pathspec	`1.1.1`	indirect
PyPI	pfzy	`0.3.4`	indirect
PyPI	pillow	`12.3.0`	indirect
PyPI	platformdirs	`4.11.1`	indirect
PyPI	pluggy	`1.6.0`	indirect
PyPI	prettytable	`3.18.0`	indirect
PyPI	priority	`2.0.0`	indirect
PyPI	prometheus-client	`0.26.0`	indirect
PyPI	prometheus-fastapi-instrumentator	`7.0.0`	indirect
PyPI	prompt-toolkit	`3.0.53`	indirect
PyPI	propcache	`0.5.2`	indirect
PyPI	proto-plus	`1.28.3`	indirect
PyPI	protobuf	`7.35.1`	indirect
PyPI	psutil	`7.2.2`	indirect
PyPI	pulp	`3.3.2`	indirect
PyPI	py-spy	`0.4.2`	indirect
PyPI	pyarrow	`25.0.0`	indirect
PyPI	pyasn1	`0.6.4`	indirect
PyPI	pyasn1-modules	`0.4.2`	indirect
PyPI	pybind11	`—`	indirect
PyPI	pybind11	`3.1.0`	indirect
PyPI	pycodestyle	`2.12.1`	indirect
PyPI	pycparser	`3.0`	indirect
PyPI	pydantic	`2.13.4`	indirect
PyPI	pydantic-core	`2.46.4`	indirect
PyPI	pydata-sphinx-theme	`0.20.0`	indirect
PyPI	pyflakes	`3.2.0`	indirect
PyPI	pygithub	`2.9.1`	indirect
PyPI	pygments	`2.20.0`	indirect
PyPI	pyjwt	`2.13.0`	indirect
PyPI	pylint	`3.2.6`	indirect
PyPI	pynacl	`1.6.2`	indirect
PyPI	pyre-extensions	`0.0.32`	indirect
PyPI	pytest	`9.1.1`	indirect
PyPI	pytest-asyncio	`1.4.0`	indirect
PyPI	pytest-cov	`7.1.0`	indirect
PyPI	pytest-mock	`3.15.1`	indirect
PyPI	pytest-random-order	`1.2.0`	indirect
PyPI	python-dateutil	`2.9.0.post0`	indirect
PyPI	python-discovery	`1.5.1`	indirect
PyPI	python-gitlab	`8.5.0`	indirect
PyPI	python-multipart	`0.0.32`	indirect
PyPI	pytokens	`0.4.1`	indirect
PyPI	pytz	`2026.3.post1`	indirect
PyPI	pywin32	`312`	indirect
PyPI	pyyaml	`6.0.3`	indirect
PyPI	pyyaml-ft	`8.0.0`	indirect
PyPI	quack-kernels	`0.4.1`	indirect
PyPI	quart	`0.21.0`	indirect
PyPI	rapidyaml	`0.15.2`	indirect
PyPI	ray	`2.56.1`	indirect
PyPI	referencing	`0.37.0`	indirect
PyPI	regex	`2026.7.19`	indirect
PyPI	requests	`—`	indirect
PyPI	requests	`2.34.2`	indirect
PyPI	requests-toolbelt	`1.0.0`	indirect
PyPI	rich	`15.0.0`	indirect
PyPI	roman-numerals	`4.1.0`	indirect
PyPI	rpds-py	`2026.6.3`	indirect
PyPI	ruff	`0.9.10`	indirect
PyPI	s3fs	`2026.6.0`	indirect
PyPI	safetensors	`0.8.0`	indirect
PyPI	scipy	`1.18.0`	indirect
PyPI	sentencepiece	`—`	indirect
PyPI	sentencepiece	`0.2.2`	indirect
PyPI	sentry-sdk	`2.66.1`	indirect
PyPI	setuptools	`—`	indirect
PyPI	setuptools	`84.0.0`	indirect
PyPI	shellingham	`1.5.4`	indirect
PyPI	six	`1.17.0`	indirect
PyPI	slack-sdk	`3.43.0`	indirect
PyPI	smart-open	`8.0.1`	indirect
PyPI	sniffio	`1.3.1`	indirect
PyPI	snowballstemmer	`3.1.1`	indirect
PyPI	soupsieve	`2.9.2`	indirect
PyPI	sphinx	`9.1.0`	indirect
PyPI	sphinx-autobuild	`2025.8.25`	indirect
PyPI	sphinx-autodoc2	`0.5.0`	indirect
PyPI	sphinx-copybutton	`0.5.2`	indirect
PyPI	sphinxcontrib-applehelp	`2.0.0`	indirect
PyPI	sphinxcontrib-devhelp	`2.0.0`	indirect
PyPI	sphinxcontrib-htmlhelp	`2.1.0`	indirect
PyPI	sphinxcontrib-jsmath	`1.0.1`	indirect
PyPI	sphinxcontrib-qthelp	`2.0.0`	indirect
PyPI	sphinxcontrib-serializinghtml	`2.0.0`	indirect
PyPI	starlette	`0.52.1`	indirect
PyPI	sympy	`1.14.0`	indirect
PyPI	tabulate	`0.10.0`	indirect
PyPI	tensorboard	`2.21.0`	indirect
PyPI	tensorboard-data-server	`0.7.2`	indirect
PyPI	tensorflow	`—`	indirect
PyPI	tensorstore	`—`	indirect
PyPI	tensorstore	`0.1.84`	indirect
PyPI	tensorstore	`0.1.85`	indirect
PyPI	tiktoken	`0.13.0`	indirect
PyPI	tilelang	`0.1.8`	indirect
PyPI	tokenizers	`0.22.2`	indirect
PyPI	toml	`0.10.2`	indirect
PyPI	tomlkit	`0.15.1`	indirect
PyPI	torch-c-dlpack-ext	`0.1.5`	indirect
PyPI	torchx	`0.7.0`	indirect
PyPI	tqdm	`—`	indirect
PyPI	tqdm	`4.67.3`	indirect
PyPI	tqdm	`4.70.0`	indirect
PyPI	transformer-engine	`2.18.0+e7c550c5`	indirect
PyPI	transformers	`—`	indirect
PyPI	transformers	`5.14.1`	indirect
PyPI	triton	`3.7.1`	indirect
PyPI	trove-classifiers	`2026.6.1.19`	indirect
PyPI	typer	`0.27.1`	indirect
PyPI	typing-extensions	`4.16.0`	indirect
PyPI	typing-inspect	`0.9.0`	indirect
PyPI	typing-inspection	`0.4.3`	indirect
PyPI	tzdata	`2025.3`	indirect
PyPI	urllib3	`2.7.0`	indirect
PyPI	uvicorn	`0.52.1`	indirect
PyPI	virtualenv	`21.7.3`	indirect
PyPI	wandb	`0.28.1`	indirect
PyPI	watchfiles	`1.2.0`	indirect
PyPI	wcmatch	`10.2.1`	indirect
PyPI	wcwidth	`0.8.2`	indirect
PyPI	webdataset	`1.0.2`	indirect
PyPI	websockets	`17.0.1`	indirect
PyPI	werkzeug	`3.1.8`	indirect
PyPI	wget	`3.2`	indirect
PyPI	win32-setctime	`1.2.0`	indirect
PyPI	wrapt	`2.3.0`	indirect
PyPI	wsproto	`1.3.2`	indirect
PyPI	xattr	`1.3.0`	indirect
PyPI	xxhash	`3.8.1`	indirect
PyPI	yarl	`1.24.5`	indirect
PyPI	z3-solver	`4.15.4.0`	indirect
PyPI	zipp	`4.1.0`	indirect
PyPI	zstandard	`0.25.0`	indirect

Dependency advisories 0

Installing pypi:megatron-core@0.18.2 pulls in 3 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.

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 v0.31.0 — full methodology · metrics wiki.

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

Software health index

Score profile

Ownership

Package ecosystems

Metrics by category

Key facts

Data collection warnings

More detail

Feedback

Related records