Public record
Software health reportschema 0.34.0 · metrics 2.10.0 · 2026-08-28 06:02 UTC

Blaizzy / mlx-vlm

MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) on your Mac using MLX.

PythonMIT★ 5,431 stars⑂ 748 forkssince Apr 2024View on GitHub ↗
KindCommand-line toolLibraryhow this is determined
#llava#llm#mlx#vision-transformer#apple-silicon#idefics#local-ai#paligemma#vision-framework#vision-language-model#florence2#molmo#pixtral

Blaizzy/mlx-vlm holds a health index of 95 out of 100, placing it in the Exceptional band. It scores highest on Vitality (98/100) and lowest on Security (60/100). It was last updated today. 2 contributors account for most of its recent work.

95
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.

95
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 82 is calibrated to 95 on the published index scale (record calibration 2026-08-02).

Ownership

Prince CanumaPersonal account
1,494 followers106 public repossince Nov 2016kulissiwa.com

This repository is owned by a personal account. A single-owner project carries more continuity risk than an organization-backed one.

Package ecosystems

RegistryPackageVersionDownloads / moVersionsLast publish
PyPImlx-vlm0.6.17881,222801 day ago

Metrics by category

Vitality

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

98Exceptional · 21% of overall

Development activity

96Exceptional
How it's scored
36/36Push recency — last push 0 days ago
31.8/36Commit cadence — 46/52 weeks with commits
18/18Commit volume — 1,039 commits in the last year
10/10OpenSSF Scorecard: Maintained — 30 commit(s) and 8 issue activity found in the last 90 days -- score normalized to 10
Inputs used
commits_last_year1,039
human_commit_share1
days_since_last_push0
active_weeks_last_year46

Release discipline

100Exceptional
How it's scored
27/27Ships releases — 80 releases published
36/36Release recency — latest release 1 days ago
27/27Release cadence — a release every ~3.8 days
0/10OpenSSF Scorecard: Signed-Releases — no data
Inputs used
releases_count80
latest_release_tagv0.6.17
releases_from_tagsno
days_since_latest_release1
mean_days_between_releases3.8
Excluded from scoring (no data or not applicable): OpenSSF Scorecard: Signed-Releases. Remaining weights renormalized.

Community & Adoption

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

86Excellent · 17% of overall

Popularity & adoption

92Excellent
How it's scored
60/60Stars — 5,431 stars
23.9/25Forks — 748 forks
8.5/15Watchers — 35 watchers
Inputs used
forks748
stars5,431
watchers35
growth_stateunverified
growth_factor_pct100
growth_unverified_reasonno_history

Community health

70Good
How it's scored
22.5/22.5README
22.5/22.5License — recognized license (MIT)
18/18CONTRIBUTING guide
0/13.5Code of conduct
0/7.2Issue template
0/6.3PR template
Inputs used
has_readmeyes
has_licenseyes
readme_badges1
has_contributingyes
has_issue_templateno
has_code_of_conductno
readme_badge_servicesgithub.com
has_pull_request_templateno

Ecosystem adoption (downloads)

99Exceptional
How it's scored
79.3/80Monthly downloads — 881,222 downloads/month across pypi
0/20Registry dependents — not reported by this ecosystem
Inputs used
packagesmlx-vlm
dependents—
ecosystemspypi
total_downloads—
monthly_downloads881,222
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?

76Good · 23% of overall

Maintainer resilience (bus factor)

62Moderate
How it's scored
25.2/54Bus factor — 2 contributor(s) cover half of all commits
12.8/22.5Commit distribution — top contributor authored 43% of commits
13.5/13.5Contributor breadth — 100 contributors
10/10OpenSSF Scorecard: Contributors — project has 7 contributing companies or organizations
Inputs used
bus_factor2
contributors_sampled100
top_contributor_share0.43

Issue & PR responsiveness

78Good
How it's scored
36.3/42Issue resolution — 86% of issues closed
22/30PR acceptance — 867/1,181 decided PRs merged
5.2/13Newcomer PR acceptance — 2/5 first-time contributors' PRs merged in 30d
15/15OpenSSF Scorecard: Code-Review — all changesets reviewed
Inputs used
merged_prs867
open_issues106
closed_issues681
prs_merged_7d43
prs_decided_7d57
prs_merged_30d46
prs_decided_30d60
issue_closed_ratio0.865
closed_unmerged_prs314
first_time_authors_30d5
first_time_prs_merged_30d2
first_time_prs_decided_30d5

Ownership & stewardship

72Good
How it's scored
10/30Ownership backing — personal (user) account
0/20Verified domain — not applicable to user accounts
22.8/25Owner reach — 1,494 followers of Blaizzy
25/25Track record — 106 public repos, account ~9 yr old
Inputs used
followers1,494
owner_typeUser
is_verified—
owner_loginBlaizzy
public_repos106
account_age_days3,574
Excluded from scoring (no data or not applicable): Verified domain. Remaining weights renormalized.

Package maintenance

100Exceptional
How it's scored
25/25Published & resolvable — 1 package(s) on pypi
35/35Publish recency — latest publish 1 days ago
20/20Version history — 80 published versions
20/20Not deprecated — active, not deprecated or yanked
Inputs used
packagesmlx-vlm
ecosystemspypi
any_deprecatedno
min_days_since_publish1

Engineering Quality

Are baseline engineering and documentation practices in place?

88Excellent · 19% of overall

Engineering practices

90Excellent
How it's scored
24/24CI workflows — 2 workflow(s)
24/24Tests present
16/16Linter config
9.6/9.6Pre-commit hooks
0/6.4.editorconfig
16/20OpenSSF Scorecard: CI-Tests — 14 out of 16 merged PRs checked by a CI test -- score normalized to 8
Inputs used
has_ciyes
has_testsyes
has_editorconfigno
has_linter_configyes
has_precommit_configyes

Documentation

85Excellent
How it's scored
30/30README
25/25Documentation directory
0/15Documentation / homepage site
10/10Repository description
10/10Topics — 13 topics
10/10Wiki
Inputs used
topicsllava, llm, mlx, vision-transformer, apple-silicon, idefics, local-ai, paligemma, vision-framework, vision-language-model, florence2, molmo, pixtral
has_wikiyes
homepage—
docs_site—
has_readmeyes
has_docs_diryes
has_descriptionyes

Security

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

60Moderate · 16% of overall

Security posture

50Moderate
How it's scored
7.5/7.5Binary-Artifacts — no binaries found in the repo
0/7.5Branch-Protection — no data
2/2.5CI-Tests — 14 out of 16 merged PRs checked by a CI test -- score normalized to 8
0/2.5CII-Best-Practices — no effort to earn an OpenSSF best practices badge detected
7.5/7.5Code-Review — all changesets reviewed
2.5/2.5Contributors — project has 7 contributing companies or organizations
10/10Dangerous-Workflow — no dangerous workflow patterns detected
0/7.5Dependency-Update-Tool — no update tool detected
0/5Fuzzing — project is not fuzzed
2.5/2.5License — license file detected
7.5/7.5Maintained — 30 commit(s) and 8 issue activity found in the last 90 days -- score normalized to 10
5/5Packaging — packaging workflow detected
0.5/5Pinned-Dependencies — dependency not pinned by hash detected -- score normalized to 1
0/5SAST — SAST tool is not run on all commits -- score normalized to 0
0/5Security-Policy — security policy file not detected
0/7.5Signed-Releases — no data
0/7.5Token-Permissions — detected GitHub workflow tokens with excessive permissions
0/7.5Vulnerabilities — 62 existing vulnerabilities detected
Inputs used
sourceopenssf_scorecard
checks_evaluated16
scorecard_versionv5.5.0
checks_inconclusive2
scorecard_aggregate5
Excluded from scoring (no data or not applicable): Branch-Protection, Signed-Releases. Remaining weights renormalized.

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_packages50
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:mlx-vlm@0.6.17 runtime dependency closure — what installing the published package pulls in — 50 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.

64Moderate · 4% of overall

Agent context & guidance

35Weak
How it's scored
0/45Agent instructions — no CLAUDE.md / AGENTS.md / editor rules
0/15Machine-readable docs (llms.txt)
35.2/40Legible commit history — 66 of 100 human commits state their intent (structured subject or explanatory body)
Inputs used
has_llms_txtno
llms_txt_url—
legible_history_share0.66
agent_instruction_files—
agent_instruction_max_bytes—

Verify loop (build / test / typecheck)

65Good
How it's scored
0/18One-command bootstrap
22/22Automated tests
11/11Lint / format config
11/11Static type checking — mlx_vlm/py.typed
10/10Reproducible environment — lockfile
10/10Demonstrated agent practice — 18 of the last 100 commits agent-authored or agent-credited
0/8Automated maintenance — no automated dependency updates observed
1/10OpenSSF Scorecard: Pinned-Dependencies — dependency not pinned by hash detected -- score normalized to 1
Inputs used
has_nixno
has_testsyes
lockfilesuv.lock
has_dockerfileno
typed_languageno
bootstrap_files—
has_devcontainerno
has_linter_configyes
typecheck_configsmlx_vlm/py.typed
agent_commit_share0.18
toolchain_manifests—
dependency_bot_commit_share0

Code legibility for models

81Excellent
How it's scored
27/45Type-checkable code — Python with type-check config (mlx_vlm/py.typed)
54.2/55Manageable file sizes — 20/1,371 source files over 60KB
Inputs used
primary_languagePython
largest_source_bytes597,993
source_files_sampled1,371
oversized_source_files20

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, notebooks
Inputs used
example_dirsexamples, notebooks
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

5,431GitHub stars
100contributors
1,039commits, last 12 months
0days since last push
80releases
2bus factor
106open 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 15 authors (cap 12)

More detail

Star and fork history 0 ★ / 748 ⇿
0Stars
748Forks
80Releases

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.

0125250375500625750699642024-042025-062026-08
Major 0Minor 6Patch 74

Each point covers 3 days.

OpenSSF Scorecard 5.0 / 10
5.0aggregate

Independent, tool-agnostic security assessment from the open-source OpenSSF Scorecard. Each check rewards a security practice, not a specific vendor's tool. Checks Scorecard could not determine are marked n/a and excluded from the security score (never counted as zero).Scorecard v5.5.0 · 2026-08-28 06:01 UTC

10Binary-Artifactsno binaries found in the repo
n/aBranch-Protectioninternal error: error during branchesHandler.setup: internal error: some github tokens can't read classic branch protection rules: https://github.com/ossf/scorecard-action/blob/main/docs/authentication/fine-grained-auth-token.md
8CI-Tests14 out of 16 merged PRs checked by a CI test -- score normalized to 8
0CII-Best-Practicesno effort to earn an OpenSSF best practices badge detected
10Code-Reviewall changesets reviewed
10Contributorsproject has 7 contributing companies or organizations
10Dangerous-Workflowno dangerous workflow patterns detected
0Dependency-Update-Toolno update tool detected
0Fuzzingproject is not fuzzed
10Licenselicense file detected
10Maintained30 commit(s) and 8 issue activity found in the last 90 days -- score normalized to 10
10Packagingpackaging workflow detected
1Pinned-Dependenciesdependency not pinned by hash detected -- score normalized to 1
0SASTSAST tool is not run on all commits -- score normalized to 0
0Security-Policysecurity policy file not detected
n/aSigned-Releasesno releases found
0Token-Permissionsdetected GitHub workflow tokens with excessive permissions
0Vulnerabilities62 existing vulnerabilities detected
All dependencies 113

Full resolved dependency set from the GitHub dependency graph: 0 direct and 113 indirect (transitive) packages. The transitive closure is complete when the repository commits a lockfile.

RegistryPackageVersionRelation
PyPIaiohappyeyeballs2.6.1indirect
PyPIaiohttp3.13.3indirect
PyPIaiosignal1.4.0indirect
PyPIannotated-doc0.0.4indirect
PyPIannotated-types0.7.0indirect
PyPIanyio4.12.1indirect
PyPIasync-timeout5.0.1indirect
PyPIattrs25.4.0indirect
PyPIaudioop-lts0.2.2indirect
PyPIbrotli1.2.0indirect
PyPIcertifi2026.1.4indirect
PyPIcffi2.0.0indirect
PyPIcharset-normalizer3.4.4indirect
PyPIclick8.3.1indirect
PyPIcolorama0.4.6indirect
PyPIdatasets4.5.0indirect
PyPIdill0.4.0indirect
PyPIexceptiongroup1.3.1indirect
PyPIfastapi—indirect
PyPIfastapi0.136.3indirect
PyPIfilelock3.20.3indirect
PyPIfrozenlist1.8.0indirect
PyPIfsspec2025.10.0indirect
PyPIgradio6.15.2indirect
PyPIgradio-client2.5.0indirect
PyPIgroovy0.1.2indirect
PyPIh110.16.0indirect
PyPIhf-gradio0.4.1indirect
PyPIhf-xet1.4.3indirect
PyPIhttpcore1.0.9indirect
PyPIhttpx0.28.1indirect
PyPIhuggingface-hub1.10.2indirect
PyPIidna3.11indirect
PyPIjinja23.1.6indirect
PyPIllguidance—indirect
PyPIllguidance1.7.5indirect
PyPImarkdown-it-py4.0.0indirect
PyPImarkupsafe3.0.3indirect
PyPImdurl0.1.2indirect
PyPIminiaudio—indirect
PyPIminiaudio1.61indirect
PyPImlx—indirect
PyPImlx0.32.0indirect
PyPImlx-audio—indirect
PyPImlx-audio0.4.3indirect
PyPImlx-cpu0.30.4indirect
PyPImlx-cuda0.30.0indirect
PyPImlx-lm0.31.3indirect
PyPImlx-metal0.32.0indirect
PyPImlx-whisper—indirect
PyPImultidict6.7.1indirect
PyPImultiprocess0.70.18indirect
PyPInumpy—indirect
PyPInumpy2.2.6indirect
PyPInumpy2.4.1indirect
PyPInvidia-cublas-cu1212.9.1.4indirect
PyPInvidia-cuda-nvrtc-cu1212.9.86indirect
PyPInvidia-cudnn-cu129.18.1.3indirect
PyPInvidia-nccl-cu122.29.2indirect
PyPIopencv-python—indirect
PyPIopencv-python4.13.0.90indirect
PyPIorjson3.11.5indirect
PyPIpackaging26.0indirect
PyPIpandas2.3.3indirect
PyPIpandas3.0.0indirect
PyPIpillow—indirect
PyPIpillow12.1.0indirect
PyPIpropcache0.4.1indirect
PyPIprotobuf6.33.4indirect
PyPIpyarrow23.0.0indirect
PyPIpycparser3.0indirect
PyPIpydantic2.12.5indirect
PyPIpydantic-core2.41.5indirect
PyPIpydub0.25.1indirect
PyPIpygments2.19.2indirect
PyPIpython-dateutil2.9.0.post0indirect
PyPIpython-multipart—indirect
PyPIpython-multipart0.0.22indirect
PyPIpytz2025.2indirect
PyPIpyyaml6.0.3indirect
PyPIregex2026.1.15indirect
PyPIrequests—indirect
PyPIrequests2.32.5indirect
PyPIrich14.3.1indirect
PyPIsafehttpx0.1.7indirect
PyPIsafetensors0.8.0indirect
PyPIscipy1.15.3indirect
PyPIscipy1.17.1indirect
PyPIsemantic-version2.10.0indirect
PyPIsentencepiece—indirect
PyPIsentencepiece0.2.1indirect
PyPIshellingham1.5.4indirect
PyPIsix1.17.0indirect
PyPIsounddevice0.5.5indirect
PyPIspeechrecognition—indirect
PyPIstarlette—indirect
PyPIstarlette1.2.0indirect
PyPItokenizers0.22.2indirect
PyPItomlkit0.13.3indirect
PyPItqdm—indirect
PyPItqdm4.67.1indirect
PyPItransformers—indirect
PyPItransformers5.14.0indirect
PyPItyper0.21.1indirect
PyPItyping-extensions4.15.0indirect
PyPItyping-inspection0.4.2indirect
PyPItzdata2025.3indirect
PyPIurllib32.6.3indirect
PyPIuvicorn—indirect
PyPIuvicorn0.40.0indirect
PyPIwebsockets—indirect
PyPIxxhash3.6.0indirect
PyPIyarl1.22.0indirect
Dependency advisories 0

Installing pypi:mlx-vlm@0.6.17 pulls in 50 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 Blaizzy/mlx-vlm. Read them side by side →

94raullenchai/Rapid-MLXThe fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separation, cloud routing. Drop-in OpenAI replacement. Works with Claude Code, Cursor, Aider.Exceptional96MakazhanAlpamys/SoupFine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.Exceptional95Blaizzy/mlx-audioA text-to-speech (TTS), speech-to-text (STT) and speech-to-speech (STS) library built on Apple's MLX framework, providing efficient speech analysis on Apple Silicon.Exceptional95maziyarpanahi/openmedLocal-first healthcare AI: clinical NER & HIPAA PII de-identification that runs 100% on-device. 2,200+ medical models, 21 languages, Apple MLX + Python, no cloud, no patient data leaving your network. Apache-2.0Exceptional95qualcomm/GenieXRun frontier LLMs and VLMs locally on Qualcomm devices across NPU, GPU, and CPU with a few lines of codeExceptional94libre-webui/libre-webuiA local-first workspace for chat, private knowledge, artifacts, and isolated model-driven work. Self-hosted. Provider-flexible. Apache 2.0.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.