vllm-project/llm-compressor erreicht einen Gesundheitsindex von 84 von 100 und liegt damit im Bereich Gut. Am stärksten schneidet es bei Vitality (95/100) ab, am schwächsten bei AI Readiness (46/100). Zuletzt heute aktualisiert. 5 Mitwirkende tragen den Großteil der jüngsten Arbeit.
84
gesamt / 100
Gut
Software-Gesundheitsindex
Metriken werden auf einer Skala von 1–100 in gewichtete Kategorien gruppiert. Der Gesamtwert beginnt als ihr Mittel; sobald öffentliche Evidenz die Richtlinie für Hochrisikojurisdiktionen auslöst, wird die Bewertung angepasst und erhält die Obergrenze 49 (Gefährdet). AI Readiness liegt außerhalb.
84
Exzellent85-100Vorbildlich; erfüllt im Wesentlichen alle geprüften Kriterien
Gut70-84Gesund; geringfügige Lücken
Mittel50-69Akzeptabel mit deutlichen Lücken; Überprüfung empfohlen
Gefährdet30-49Erhebliche Schwächen; eine Übernahme erfordert Vorsicht
Kritisch1-29Schwerwiegende Probleme (aufgegeben, nur ein Maintainer, keine Hygiene)
Bewertungsprofil
Jede Achse ist eine Kategorie. Die Form zählt mehr als der Durchschnitt — ein gesundes Projekt füllt die gesamte Fläche, während ein Profil aus Spitzen und Kratern bedeutet, dass Stärke in einer Dimension Risiken in einer anderen verdeckt.
Dieses Repository wird von einer Organisation getragen — geteilte, rechenschaftspflichtige Trägerschaft, die jeden einzelnen Maintainer überdauern kann.
Direkte Abhängigkeiten ohne bekannte Advisories — 2 betroffen: pillow 12.2.0 (critical 9.1), torch 2.12.0 (moderate 5.3)
25/25
Indirekte Abhängigkeiten ohne bekannte Advisories — keine indirekte Abhängigkeit trägt ein bekanntes Advisory
35.3/40
Keine offenen Advisories — 1 Paket(e) mit Advisory seit über 90 Tagen unbehandelt; ältestes vor 480 Tagen veröffentlicht
Verwendete Eingangsdaten
source
osv
advisories
27
affected_packages
2
assessed_packages
59
unassessed_packages
0
affected_by_severity
critical 1, moderate 1
direct_affected_packages
2
Abgeglichen wurde die Laufzeit-Abhängigkeitshülle von pypi:llmcompressor@0.12.0 — das, was die Installation des veröffentlichten Pakets nach sich zieht — mit 59 Paketen. Erreichbarkeit wird nicht analysiert.
Wie gut ist das Repository dafür ausgestattet, mit KI-Coding-Agenten entwickelt und gepflegt zu werden? Ein unabhängiges, experimentelles Badge — Gewicht 0,0, es wird eigenständig ausgewiesen und verändert den Gesamt-Gesundheitswert nicht.
Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token
Weitere Details
Stern- und Fork-Verlauf 0 ★ / 586 ⇿
0Sterne
586Forks
27Releases
Wann jeder Stern und Fork hinzugefügt wurde, von GitHub erfasst und nach Tagen gruppiert. Das kumulierte Wachstum steht direkt über den täglichen Zugängen, aus denen es besteht, sodass beide gegeneinander lesbar sind: stetiger organischer Zuwachs sieht ganz anders aus als ein abrupter, kurzlebiger Ausschlag. Wo dieser Unterschied messbar ist, wird er als Wachstumsauthentizität ausgewiesen.
Major 0Minor 12Patch 6
Jeder Punkt umfasst 2 Tage.
OpenSSF Scorecard 5.9 / 10
5.9Gesamtwert
Unabhängige, werkzeugneutrale Sicherheitsbewertung durch das quelloffene OpenSSF Scorecard. Jede Prüfung honoriert eine Sicherheits-Praxis, nicht das Werkzeug eines bestimmten Anbieters. Prüfungen, die Scorecard nicht ermitteln konnte, sind mit k. A. markiert und vom Sicherheitswert ausgeschlossen (nie als null gezählt).Scorecard v5.5.0 · 2026-07-25 06:48 UTC
Vollständig aufgelöster Abhängigkeitssatz aus dem GitHub-Abhängigkeitsgraphen: 0 direkte und 1 indirekte (transitive) Pakete. Die transitive Hülle ist vollständig, wenn das Repository eine Lockfile eincheckt.
Registry
Paket
Version
Beziehung
PyPI
setuptools-scm
8.2.0
indirekt
Abhängigkeits-Advisories 2
Die Installation von pypi:llmcompressor@0.12.0 zieht 59 Pakete nach sich, direkt und transitiv: 2 tragen bekannte Advisories, davon 2 direkte Abhängigkeiten.
Paket
Version
Beziehung
Schweregrad
Advisories
Behoben in
pillow
12.2.0
direkt
kritisch
26
12.3.0
torch
2.12.0
direkt
mittel
1
2.13.0
Ein Advisory bedeutet, dass die im Abhängigkeitsgraphen erfasste Version in den betroffenen Bereich eines Advisories fällt. Erreichbarkeit wird nicht analysiert, und der Graph enthält Entwicklungs- und Test-Pins — ein Fund kann das Werkzeug betreffen und nicht die ausgelieferte Software.
JSON-Rohbericht maschinenlesbar
{
"data": {
"repo": {
"topics": [
"compression",
"quantization"
],
"is_fork": false,
"size_kb": 40041,
"has_wiki": true,
"homepage": "https://docs.vllm.ai/projects/llm-compressor",
"languages": {
"Shell": 7207,
"Python": 1562746,
"Makefile": 1772
},
"pushed_at": "2026-07-24T23:22:47Z",
"created_at": "2024-06-20T20:13:34Z",
"owner_type": "Organization",
"updated_at": "2026-07-25T03:46:32Z",
"description": "Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM",
"is_archived": false,
"is_disabled": false,
"license_spdx": "Apache-2.0",
"default_branch": "main",
"license_spdx_raw": "Apache-2.0",
"primary_language": "Python",
"significant_languages": [
"Python"
]
},
"owner": {
"blog": null,
"name": "vLLM",
"type": "Organization",
"login": "vllm-project",
"company": null,
"location": null,
"followers": 3752,
"avatar_url": "https://avatars.githubusercontent.com/u/136984999?v=4",
"created_at": "2023-06-18T14:54:57Z",
"is_verified": null,
"public_repos": 45,
"account_age_days": 1132
},
"license": {
"state": "standard",
"spdx_id": "Apache-2.0",
"raw_spdx": "Apache-2.0",
"file_present": true,
"scorecard_found": true,
"profile_has_license": true
},
"activity": {
"releases": [
{
"tag": "0.7.1.3",
"kind": "other",
"published_at": "2026-06-26T00:26:26Z"
},
{
"tag": "0.12.0",
"kind": "minor",
"published_at": "2026-06-15T13:02:02Z"
},
{
"tag": "0.11.0",
"kind": "minor",
"published_at": "2026-06-02T17:19:03Z"
},
{
"tag": "0.7.1.2",
"kind": "other",
"published_at": "2026-05-06T20:04:46Z"
},
{
"tag": "0.9.0.3",
"kind": "other",
"published_at": "2026-05-05T14:09:46Z"
},
{
"tag": "0.10.0.2",
"kind": "other",
"published_at": "2026-05-01T13:22:21Z"
},
{
"tag": "0.10.0.1",
"kind": "other",
"published_at": "2026-03-13T14:41:51Z"
},
{
"tag": "0.7.1.1",
"kind": "other",
"published_at": "2026-03-04T23:50:19Z"
},
{
"tag": "0.10.0",
"kind": "minor",
"published_at": "2026-03-02T16:23:13Z"
},
{
"tag": "0.9.0.2",
"kind": "other",
"published_at": "2026-02-13T22:50:21Z"
},
{
"tag": "0.9.0.1",
"kind": "other",
"published_at": "2026-01-21T18:48:17Z"
},
{
"tag": "0.9.0",
"kind": "minor",
"published_at": "2025-12-17T17:24:27Z"
},
{
"tag": "0.8.1",
"kind": "patch",
"published_at": "2025-10-08T02:13:59Z"
},
{
"tag": "0.8.0",
"kind": "minor",
"published_at": "2025-10-03T12:24:52Z"
},
{
"tag": "0.7.1",
"kind": "patch",
"published_at": "2025-08-21T21:37:14Z"
},
{
"tag": "0.7.0",
"kind": "minor",
"published_at": "2025-08-20T21:40:17Z"
},
{
"tag": "0.6.0.1",
"kind": "other",
"published_at": "2025-07-28T19:05:12Z"
},
{
"tag": "0.6.0",
"kind": "minor",
"published_at": "2025-06-24T15:22:12Z"
},
{
"tag": "0.5.2",
"kind": "patch",
"published_at": "2025-06-24T01:47:49Z"
},
{
"tag": "0.5.1",
"kind": "patch",
"published_at": "2025-04-29T01:34:26Z"
},
{
"tag": "0.5.0",
"kind": "minor",
"published_at": "2025-04-03T13:23:53Z"
},
{
"tag": "0.4.1",
"kind": "patch",
"published_at": "2025-02-20T13:21:14Z"
},
{
"tag": "0.4.0",
"kind": "minor",
"published_at": "2025-01-16T03:12:58Z"
},
{
"tag": "0.3.1",
"kind": "patch",
"published_at": "2024-12-12T13:25:28Z"
},
{
"tag": "0.3.0",
"kind": "minor",
"published_at": "2024-11-13T05:22:10Z"
},
{
"tag": "0.2.0",
"kind": "minor",
"published_at": "2024-09-23T22:24:43Z"
},
{
"tag": "0.1.0",
"kind": "minor",
"published_at": "2024-08-12T15:37:11Z"
}
],
"recent_commits": [
{
"oid": "e5eb138215726a8d00da0b308c036f77eee674f9",
"body": "## Add to Key Models\n* Add DSV4, GLM5.2, HY_V3 to key models docs\n\n## Readme\n* Fix alignment of logo and name\n\n### Before: \n<img width=\"353\" height=\"114\" alt=\"Screenshot 2026-07-24 at 10 16 31\"\nsrc=\"https://github.com/user-attachments/assets/c6556770-76e0-478b-b12a-15469aef0566\"\n/>\n\n### Now: \n<img w\n[…]\nizing models for deployment with vLLM,\nincluding:\"\n\n* Add notes about low bit quantization and reap\n\n## Bugfix\n\n* Fix bug in glm5.2 example\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Docs] Add DSV4, GLM5.2, HY_V3 to key models, Update README (#2967)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-24T16:29:38Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c4fa85b8ae38498981a85e31fbdcd067fc00f4e4",
"body": "## Prerequisites ##\n### Accelerate/ Transformers ###\n- [x] https://github.com/huggingface/accelerate/pull/4083\n- [x] https://github.com/huggingface/transformers/pull/46749\n- [x] https://github.com/huggingface/transformers/pull/46849\n- [x] https://github.com/huggingface/transformers/pull/47018\n - [x\n[…]\n====\nHello my name is Inigo Montoya. You killed my father. Prepare to die<｜end▁of▁sentence｜>\n==========================================\n```\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[DSV4] DeepSeekV4 Pro (#2858)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-24T16:17:07Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "8716438365b0b8a6169dd0251743887b02ac7d4b",
"body": "…862) (#2912)\n\n## Summary\n\nFixes #2862 - two failures when running GPTQ on GLM-5.2 (`glm_moe_dsa`):\n\n- **OOM with default targets**: whole `GlmMoeDsaDecoderLayer` is one\nsubgraph holding all 128+ experts → all Hessians on GPU simultaneously.\n- **fx-trace failure with `[\"Linear\"]`**: forces tracer in\n[…]\narpujar@gmail.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>\nCo-authored-by: Brian Dellabetta <bdellabe@redhat.com>",
"is_bot": false,
"headline": "fix(glm_moe_dsa): correct sequential_targets for GPTQ calibration (#2…",
"author_name": "Jayakumar Pujar",
"author_login": "jayakumarpujar",
"committed_at": "2026-07-24T15:53:11Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "887b9762bd65239d6c631bce9ffbf686257349d1",
"body": "SUMMARY:\n\nUpdate failing test with a monkey-patched smoke test model with fp_quant\nmethod\n```\n| Root Cause | Jobs Affected | Type |\n| - | - | - |\n| `test_oneshot_rejects_pre_quantized_smoke_model` fails with `RuntimeError` instead of expected `ValueError` when loading `google/gemma-4-E2B-it-qat-mobi\n[…]\nwdc-k8s-h100-solo, Python 3.10) | Code Regression (PR #2909) |\n```\n\n\nTEST PLAN:\ntest succeeds on any hardware/transformers version now\n\n---------\n\nSigned-off-by: Brian Dellabetta <bdellabe@redhat.com>",
"is_bot": false,
"headline": "[testfix] update small quant model for test (#2968)",
"author_name": "Brian Dellabetta",
"author_login": "brian-dellabetta",
"committed_at": "2026-07-24T15:33:50Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "3f8f2e10a78a4050eefc021da62ed994a34e7952",
"body": "SUMMARY:\n- Update what's new section\n- Fix some inconsistencies \n- Update docs homepage with latest release notes\n\nSigned-off-by: Dipika <dipikasikka1@gmail.com>",
"is_bot": false,
"headline": "[Docs] Update what's new; fix some docs (#2964)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-24T14:40:40Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "397dfe8f806a5a25b9393621f4573100f7d10beb",
"body": "SUMMARY:\n- Fix existing links\n- Add governance docs",
"is_bot": false,
"headline": "[Docs] Add governance guidelines and fix links (#2963)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-24T14:40:06Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "dc318481793116f2e60add1517a6df721751ee6a",
"body": "With some more details and relevant points",
"is_bot": false,
"headline": "Update `CONTRIBUTING` doc (#2962)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-24T14:24:56Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "26bb828b23c3242fd0f5dff35393ca31fb0da161",
"body": "SUMMARY:\nAdded NanbeigeForCausalLM to the AWQ static mapping registry. Nanbeige's\ndecoder layout uses input_layernorm, q/k/v/o attention projections,\npost_attention_layernorm, and gate/up/down MLP projections, which\nmatches the default AWQ mapping pattern. So this PR simply adds the\nmapping explicit\n[…]\ns/llmcompressor/modifiers/transform/awq/test_dynamic_mappings.py\n- pytest\ntests/llmcompressor/modifiers/transform/awq/test_dynamic_mappings.py\n\nCo-authored-by: f-baig <f-baig@users.noreply.github.com>",
"is_bot": false,
"headline": "Add Nanbeige AWQ mappings (#2966)",
"author_name": "frhnbaig",
"author_login": "f-baig",
"committed_at": "2026-07-24T13:21:43Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e531e9296bb7bf673100d27d1b05be809713db4e",
"body": "PREREQUISITES:\n\n* https://github.com/vllm-project/compressed-tensors/pull/779\n\nSUMMARY:\n- Fixes #2907 by adding an early `oneshot` validation step for models\nthat already carry quantization metadata.\n- Uses `compressed_tensors.utils.getattr_chain` inline in the `Oneshot`\ninitialization path, avoidin\n[…]\nly.github.com>\nSigned-off-by: Brian Dellabetta <bdellabe@redhat.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>\nCo-authored-by: Brian Dellabetta <bdellabe@redhat.com>",
"is_bot": false,
"headline": "[Oneshot] validate for pre-quantized models as inputs (#2909)",
"author_name": "w3lld1",
"author_login": "w3lld1",
"committed_at": "2026-07-23T20:18:01Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "5094ca72c1ee28777e7dd4c49f6303e31d981344",
"body": "…tistics (#2829)\n\n## Purpose ##\n* Add warnings for when/if calibration goes wrong\n\n## Changes ##\n* Warn user when they attempt to use an observer to calibrate a module\nwithout having first observed statistics. Because the sequential and\nbasic pipelines always try to calibrate all modules in the mode\n[…]\n triggered when\n`moe_calibration_context` is not used on moe models\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: HDCharles <39544797+HDCharles@users.noreply.github.com>",
"is_bot": false,
"headline": "Add warning for attempting to calibrate modules which do not have sta…",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-23T19:44:43Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "9e953aaba129a3406f0cfc26a0b2e253ade67233",
"body": "## Purpose ##\n* Support GLM5.2\n\n## Prerequisites ##\n### Accelerate/ Transformers ###\n- [x] https://github.com/huggingface/accelerate/pull/4083\n- [x] https://github.com/huggingface/transformers/pull/46749\n- [x] https://github.com/huggingface/transformers/pull/46849\n- [x] https://github.com/huggingfac\n[…]\n#\n- [x] https://github.com/vllm-project/vllm/pull/47780\n\n## Testing ##\n* Used to quantize https://huggingface.co/RedHatAI/GLM-5.2-FP8-NVFP4\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Example] GLM5.2 Example (#2869)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-23T19:26:05Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "7fd9a08a6a8f05d608bf3eb9224a3b9e56063773",
"body": "…ore fetch (#2813)\n\nfixes #2790 \n\ni was able to narrow it down to indeed the pin_memory causing the CUDA\nOOM issue. this is a specially case since its nested with the\ndispatcher.\n\n\nthe script im using to benchmark speed timings\n\n```python\n\n\"\"\"\nRecipe: W4A16 GPTQ (4-bit weights, 16-bit activations)\nS\n[…]\nauthored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>\nCo-authored-by: Brian Dellabetta <bdellabe@redhat.com>\nCo-authored-by: HDCharles <39544797+HDCharles@users.noreply.github.com>",
"is_bot": false,
"headline": "[IntermediatesCache] Move call to pin_memory from offload to just bef…",
"author_name": "JINO ROHIT",
"author_login": "JINO-ROHIT",
"committed_at": "2026-07-23T14:19:39Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "bf87c4dea725e7662e1d2a9e3637c5861bf93b5f",
"body": "* Better vlm support\n* Better mtp handling\n* Better uploading outside of rh team\n* Better initialization\n* Better saving\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Claude] `create-tiny-model` Improvements (#2942)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-23T05:13:16Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "42a1b830839b44f8c77d5d558baed5a3806efd04",
"body": "## Purpose ##\n* Support `HY_V3` model\n\n## Prerequisites ##\n### Transformers ###\n- [ ] https://github.com/huggingface/transformers/pull/47336\n\n## Testing ##\n* Used to generate https://huggingface.co/RedHatAI/Hy3-NVFP4-FP8\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Example] [MoE] HY_V3 (#2928)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-22T23:35:01Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "655b83d1240c37920a603b747bbfb6c218f66383",
"body": "Summary:\n\nin https://github.com/vllm-project/llm-compressor/pull/2865 we made it\nso obs clean up their stats to improve memory footprint during\nquantization. However certain modifiers weren't deleting statistics on\nDDP because of the job assignment, deletion was happening on\nobs.get_qparams which on\n[…]\nser\",\"status\":\"404\"}>\nCo-authored-by: HDCharles <{\"message\":\"Not Found\",\"documentation_url\":\"https://docs.github.com/rest/users/emails#list-email-addresses-for-the-authenticated-user\",\"status\":\"404\"}>",
"is_bot": false,
"headline": "fix DDP obs stat deletion (#2943)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-07-21T14:37:35Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "878d1c1724da51fb38cc570387406df1ccffc0a1",
"body": "## Purpose ##\n* Reduce shard size for better network transfer\n* https://huggingface.co/RedHatAI/GLM-5.2-NVFP4/discussions/1\n\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>",
"is_bot": false,
"headline": "Reduce default save shard size to 20GB (#2927)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-20T06:12:18Z",
"body_truncated": false,
"is_coding_agent": true
},
{
"oid": "c7fe59170b50e80f5707496aa43f1f34668a4cec",
"body": "## Purpose ##\n* Make `create-tiny-model` more robust and efficient by fixing common\npitfalls\n\n## Changes ##\n* Transformers often does a bad job of implementing\n`model.init_weights()` and other weight initialization logic. This trips\nup the model basically every time\n * Add a step to initialize weig\n[…]\n <kylesayrs@gmail.com>\nCo-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "[Claude] Improve common issues with `create-tiny-model` skill (#2924)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-17T16:26:44Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "e9f9aefe11c143db9c339bf8b76208cd812ca2e1",
"body": "SUMMARY:\nFix the AutoRound DDP rank-local placement regression from #2844.\n\n- keep AutoRound-managed multi-GPU offload disabled for normal\n1-GPU-per-rank DDP\n- use the current rank device when deriving the AutoRound device map\n- rebuild quantize-block inputs after moving cached tensors to the\nselect\n[…]\nessor/modifiers/autoround/test_base.py \n- pytest\ntests/llmcompressor/transformers/compression/test_compression_ddp.py -k\ntest_ddp_smoke_autoround\n\n---------\n\nSigned-off-by: yiliu30 <yi4.liu@intel.com>",
"is_bot": false,
"headline": "[Autoround][DDP] Fix rank-local device placement (#2934)",
"author_name": "Yi Liu",
"author_login": "yiliu30",
"committed_at": "2026-07-16T16:49:52Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "03226aefa596fdf6a4996be0b09627bab980dfe0",
"body": "… (#2917)\n\n## Summary\n\n`TextGenerationDataset.tokenize()` calls the processor with\n`truncation=True` and `max_length` for calibration. Since\n**transformers>=4.51** this per-call truncation (and padding) is\n**persisted onto the fast tokenizer's backend**, and then serialized\ninto `tokenizer.json` whe\n[…]\ne-svg <263951916+chiptoe-svg@users.noreply.github.com>\nCo-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Preserve tokenizer truncation/padding across calibration tokenization…",
"author_name": "chiptoe-svg",
"author_login": "chiptoe-svg",
"committed_at": "2026-07-16T16:15:19Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "7faa859a51633010fea461d351a74fa25ef44526",
"body": "SUMMARY:\n\n- ~~Depends on\nhttps://github.com/vllm-project/compressed-tensors/pull/742~~\n- Depends on https://github.com/vllm-project/llm-compressor/pull/2896\n```bash\n \n# vllm ({'pretrained': 'Yi30/Qwen3-235B-A22B-Instruct-2507-W4A16-AutoRound-iters100-nsamples256-DDP2', 'tensor_parallel_size': 4, 'm\n[…]\n |0.0065|\n# | | |strict-match | 5|exact_match|↑ |0.957|± |0.0064|\n```\n\nTEST PLAN:\n\"please outline how the changes were tested\"\n\n---------\n\nSigned-off-by: yiliu30 <yi4.liu@intel.com>",
"is_bot": false,
"headline": "[Autoround][DDP] Add Qwen MoE Example (#2844)",
"author_name": "Yi Liu",
"author_login": "yiliu30",
"committed_at": "2026-07-15T17:36:07Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "48b2260612c555528fce8c859272607fe4bc61e4",
"body": "`GlmMoeDsaForCausalLM` (GLM-5.x, DSV3.2 / DeepSeek-Sparse-Attention) is\na mixed dense/MoE MLA model but was registered to `_deepseek_mappings`,\nwhose MLP balance list includes `mlp.gate`. The leading dense layers\n(`first_k_dense_replace`) have no router, so on those layers that\npattern never matches\n[…]\nline and ran\nthe tests above.\n\n---------\n\nSigned-off-by: Karthik Kothuri <karthikkothuri2009@gmail.com>\nCo-authored-by: Claude <noreply@anthropic.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[AWQ] Fix GlmMoeDsa dense+MoE mapping collapse (#2923)",
"author_name": "Karthik Kothuri",
"author_login": "KKothuri",
"committed_at": "2026-07-14T16:48:16Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "637338d08286e140f3a2e3db833ead61d71f0c3c",
"body": "…-Next AWQ (#2681)\n\n## Summary\n\nFixes #2680 - `<think>` leakage and missing `</think>` after AWQ\nquantization of Qwen3-Next thinking models.\n\n**Root cause:** `apply_chat_template` in `qwen3_next_example.py` was\ncalled without `enable_thinking=False`. Qwen3-Next thinking tokenizers\ndefault to `enable\n[…]\n799@gmail.com>\nSigned-off-by: Brian Dellabetta <bdellabe@redhat.com>\nCo-authored-by: Brian Dellabetta <bdellabe@redhat.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "fix: add enable_thinking flag and reasoning dataset example for Qwen3…",
"author_name": "Jayakumar Pujar",
"author_login": "jayakumarpujar",
"committed_at": "2026-07-13T17:29:57Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "57ed2be4b9169f07c8ee065757158bd66b0925a6",
"body": "Summary:\nto fix a memory \"leak\" we have to delete statistics, this is a bit\ncomplicated due to fusion. There's now enough functionality specific to\nfusion that we should boil it into a separate object to that observer\ncan be a bit simpler.\n\nadded fusion_handler\nadded functionality to delete statisti\n[…]\ndisabled.</sup>\n\n<!-- codesmith:autofix:disabled -->\n<!-- /codesmith:footer -->\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "observer fusion and deletion (#2865)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-07-10T20:31:28Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "74dff970deb0f1de0ef3ee6e8de0883397029103",
"body": "SUMMARY:\n\"please provide a brief summary\"\n\n- https://huggingface.co/Yi30/Qwen3-8B-AttnW2A16-MlpW4A16-G128-AutoRound\n- https://huggingface.co/Yi30/Qwen3-8B-AttnW4A16-MlpW2A16-G128-AutoRound\n\n```bash\n# vllm ({'pretrained': '/storage/yiliu7/Qwen3-8B-AttnW2A16-MlpW4A16-G128-AutoRound', 'tensor_parallel_\n[…]\n<yi4.liu@intel.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Dipika Sikka <dipikasikka1@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "[Autoround] Add support for sub-bit quantization with examples (#2895)",
"author_name": "Yi Liu",
"author_login": "yiliu30",
"committed_at": "2026-07-10T15:05:25Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c8676edd2f6cddd0be2427f034825ca0d476fa72",
"body": "Signed-off-by: yiliu30 <yi4.liu@intel.com>\n\nSUMMARY:\n\"please provide a brief summary\"\n\n\nTEST PLAN:\n\"please outline how the changes were tested\"\n\n---------\n\nSigned-off-by: yiliu30 <yi4.liu@intel.com>\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Pin auto-round to 0.14.1 (#2896)",
"author_name": "Yi Liu",
"author_login": "yiliu30",
"committed_at": "2026-07-09T16:50:03Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "3fd8f98938dd2a13e3e23b3398b84c5509dfccc5",
"body": "## Summary\n\nAdds SpinQuant support for `Cohere2MoeForCausalLM` (e.g.\n`CohereLabs/North-Mini-Code-1.0`), along with an end-to-end NVFP4\nexample.\n\n## Changes\n\n- **`modeling/moe/cohere2_moe.py`** —\n`prepare_cohere2_moe_for_spinquant`, an architecture-specific prep\nfunction. Because Cohere2MoE uses a **\n[…]\nt to the nvfp4 cutlass backend.\n\n---------\n\nSigned-off-by: ZewenShen-Cohere <zewen.shen@cohere.com>\nCo-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>",
"is_bot": false,
"headline": "Add SpinQuant support for Cohere2MoE (#2867)",
"author_name": "ZewenShen-Cohere",
"author_login": "ZewenShen-Cohere",
"committed_at": "2026-07-09T15:23:02Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "f312622841f49c0aa178791e4440bc85f7e954f9",
"body": "The check ready label check is no longer needed, as a PR cannot be\nmerged unless buildkite runs.\n\nNote, the test still runs on this PR because the GHA is a\n`pull_request_target`, meaning that the workflow triggers from the base\n(main), where the action still exists\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>\nCo-authored-by: CodeRabbit <noreply@coderabbit.ai>",
"is_bot": false,
"headline": "[GHA] Remove `ready-label-check.yaml` (#2900)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-09T10:08:57Z",
"body_truncated": false,
"is_coding_agent": true
},
{
"oid": "7ae0f67c8b475528d4bf750861d43a022661b867",
"body": "SUMMARY:\n- Stop using internal runners\n\nSigned-off-by: Dipika <dipikasikka1@gmail.com>",
"is_bot": false,
"headline": "[Runners] Update runners to all be ubuntu-latest (#2901)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-08T22:43:03Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "5a6957b084751c2170187ee8d67b9e792b42c6bb",
"body": "## Summary\n- Expand CODEOWNERS to cover previously unlisted directories\n(`modifiers/`, `transformers/`, `pipelines/`, and corresponding `tests/`\nsubdirectories)\n- Rebalance ownership across maintainers for existing entries\n\n## Test plan\n- N/A (metadata-only change)",
"is_bot": false,
"headline": "Update CODEOWNERS (#2899)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-08T20:38:32Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "634c947f208193bcfaacbf1bcbc5b49644d7e3fa",
"body": "…ifier` (#2889)\n\nSUMMARY:\n- Consolidate shared logic for model handling for NVFP4 and FP8 into a\nshared filed\n- Add logic to handle NVFP4 for the QuantizationModifier and\nGPTQModifier\n\n---------\n\nCo-authored-by: Dipika Sikka <dsikka@dgx-b200-02.mgmt.accl-001.lab.rdu2.dc.redhat.com>",
"is_bot": false,
"headline": "[Agents] Add NVFP4 skills for the `QuantizationModifier` and `GPTQMod…",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-08T10:43:55Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "887df66b11958da8ee285d01d554c1801dd9dccc",
"body": "Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>\nSigned-off-by: Dipika Sikka <ds3822@columbia.edu>\nCo-authored-by: Dipika Sikka <dipikasikka1@gmail.com>\nCo-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>",
"is_bot": false,
"headline": "[Readme] GLM5.2 What's New (#2898)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-07T20:37:19Z",
"body_truncated": false,
"is_coding_agent": true
},
{
"oid": "80a1e4a42a990ebc59793b116163e8e0744e86fb",
"body": "SUMMARY:\nAdds \\`GraniteForCausalLM\\` (IBM Granite dense decoder) to both the AWQ\nand SmoothQuant layer-mapping registries. Maps to the existing\n\\`default_mappings\\` (AWQ) / \\`DEFAULT_SMOOTHQUANT_MAPPINGS\\`\n(SmoothQuant), since the architecture uses standard Llama-style layer\nnaming: \\`input_layernor\n[…]\nAULT_SMOOTHQUANT_MAPPINGS, #\nsmoothquant/utils.py\n\\`\\`\\`\n\n---------\n\nSigned-off-by: HumphreySun98 <humphreysun98@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Add GraniteForCausalLM mappings for AWQ and SmoothQuant (#2797)",
"author_name": "Humphrey",
"author_login": "HumphreySun98",
"committed_at": "2026-07-07T19:51:02Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "8aa00ac50e5b0c4a780ec81fdac3fcdeaa67cad4",
"body": "## Purpose ##\n* Fix bad calibration dataset processing for internvl3\n * Use an actual vision dataset to calibrate\n* Add example generation for this example\n* Use NVFP4\n\n## Changes ##\n* Fix dataset processing\n* Previously, `example[\"messages\"]` was being stringified (even though\nit's a list of dicts\n[…]\nurrounded by various items, including a computer monitor, a toy, and other desk accessories.\n==========================================\n```\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Examples] [Bugfix] Fix data processing for InternVL3 (#2892)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-07T18:46:14Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e3b60d3455d637f2584ca7a9f7b7ee851a797ef2",
"body": "…cate lm_head (#2835)\n\n## Summary\n\nQuantizing a **tied** model's embedding is currently counterproductive.\nWhen embeddings are targeted, `QuantizationMixin.start_calibration`\nunties them (existing upstream behavior) so the input and output\nembeddings can be quantized independently — but for a tied m\n[…]\nre reviewed by the submitter.\n\n---------\n\nSigned-off-by: Karthik Kothuri <karthikkothuri2009@gmail.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Claude <noreply@anthropic.com>",
"is_bot": false,
"headline": "Re-tie identically-quantized input/output embeddings to avoid a dupli…",
"author_name": "Karthik Kothuri",
"author_login": "KKothuri",
"committed_at": "2026-07-07T18:41:35Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "abcbeb417f20d3092c939dcc419d3a61714a6789",
"body": "## Summary\n\nMigrate GPU test workflows (`test-check.yaml` and\n`test-check-transformers.yaml`) from GitHub Actions to Buildkite using\nthe dispatcher pattern, and remove the original GHA workflows.\n\n- **Base tests**: Python 3.10/3.13 matrix\n- **Transformers tests**: Python 3.12, per-file pytest loop, \n[…]\n `tests/`\nchanged\n- [x] Manual build with `code-coverage: true` — coverage artifacts\nuploaded and combined\n- [x] GHA GPU workflows removed\n\n---------\n\nSigned-off-by: Deepak Kumar <deepakku@redhat.com>",
"is_bot": false,
"headline": "Add Buildkite pipeline to migrate GPU test workflows from GHA (#2876)",
"author_name": "Deepak Kumar",
"author_login": "deepak-kumar-neu",
"committed_at": "2026-07-07T17:39:42Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "0a27159e6e180af8f631426db08650f19adb9e32",
"body": "## Summary\n- Remove redundant `torch.no_grad()` context manager from\n`examples/disk_offloading/deepseek_v32_example.py`\n- Remove unused `import torch` that was only needed for `torch.no_grad`\n\n## Test plan\n- [ ] Verify the example script still runs correctly without\n`torch.no_grad`\n\n🤖 Generated with\n[…]\nBrian Dellabetta <brian-dellabetta@users.noreply.github.com>\nCo-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>\nCo-authored-by: CodeRabbit <noreply@coderabbit.ai>",
"is_bot": false,
"headline": "Remove unnecessary torch.no_grad from deepseek_v32 example (#2731)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-07T16:18:31Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "5f04e6e4aa09bfab624742869eab5af3a9bcc2bf",
"body": "## Summary\n- Replace legacy `Optional[Dict[...]]` typing with Python 3.10\n`dict[...] | None`\n- Remove now-unused `typing` imports from the sparsification utility\n\n## Tests\n- `python -m ruff check --no-cache src tests examples setup.py`\n- `python -m ruff format --check --no-cache src tests examples s\n[…]\nruns)\n\nSigned-off-by: Pruthvi226 <pruthvirajpanduga22@gmail.com>\nCo-authored-by: Pruthvi226 <pruthvirajpanduga22@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Modernize ModuleSparsificationInfo typing (#2870)",
"author_name": "Panduga Pruthvi Raj",
"author_login": "Pruthvi226",
"committed_at": "2026-07-07T15:58:33Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "dd55d139fee09dce90ba7aa99d8ab4201d7e231f",
"body": "## Summary\n- Replace `Optional[X]` with `X | None` in `events/event.py` and\n`session_functions.py`\n- Remove now-unused `Optional` imports from `typing`\n- Fix import block ordering to satisfy ruff's isort check\n\n## Problem\nThe `core/` module still has legacy `Optional[...]` type hint syntax\nfrom Pyth\n[…]\n@gmail.com>\nCo-authored-by: OpenClaw Agent <agent@openclaw.local>\nCo-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "refactor: modernize core/ module with Python 3.10+ type hints (#2834)",
"author_name": "Asad Shahid",
"author_login": "AsadShahid04",
"committed_at": "2026-07-07T15:40:35Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "c7903296cff735482e13229ff136d391246822c8",
"body": "## Purpose\n\n`untie_word_embeddings` does not actually untie the input/output\nembeddings when the model is offloaded with `compressed_tensors`\n(`OffloadCache`). The two embeddings keep sharing the same storage, so\nat save time `save_pretrained` de-duplicates them and silently drops\n`lm_head.weight`.\n\n[…]\nu'), 853154880)\noutput embedding storage: (device(type='cpu'), 853171328)\nOK: embeddings are untied (separate storage)\n```\n\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Fix untie_word_embeddings not untying offloaded models (#2866)",
"author_name": "ZewenShen-Cohere",
"author_login": "ZewenShen-Cohere",
"committed_at": "2026-07-07T15:31:03Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "d0a53511a6198bbb755bc241e748b06660b4183f",
"body": "…#2890)\n\n## Purpose ##\n* Skip `test_linearize_moe` if model is not supported by local\ntransformers version\n* Skip `test_load_quantizable_moe` if model is not supported by local\ntransformers version\n* Reduce memory usage for `deepseek_v4`\n\n## Testing ##\n* Tested that all tests pass, with 3 skipped wh\n[…]\n [100%]\n====================================== 31 passed, 3 skipped, 14 warnings in 1.79s ======================================\n```\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Tests] [Bugfix] Skip MoE tests for models which cannot be imported (…",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-07T15:02:33Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "44f862a58177aecfaa17d42f955ddf1f632f0910",
"body": "SUMMARY:\n\nAdapt the Kimi K2 example for GLM 4.6\n\n\nTEST PLAN:\n\n```\npython glm_4.6_fp8_block.py \n2026-02-10T17:33:24.378844+0000 | model_free_ptq | INFO - Copying .gitattributes GLM-4.6-FP8-BLOCK/.gitattributes\n2026-02-10T17:33:24.379164+0000 | model_free_ptq | INFO - Copying README.md GLM-4.6-FP8-BLO\n[…]\nded to: https://huggingface.co/mgoin/GLM-4.6-FP8-BLOCK\n\nSigned-off-by: mgoin <mgoin64@gmail.com>\nCo-authored-by: Dipika Sikka <dipikasikka1@gmail.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Add model_free_ptq example for glm 4.6 block fp8 (#2343)",
"author_name": "Michael Goin",
"author_login": "mgoin",
"committed_at": "2026-07-06T17:55:28Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "8cce3befe29f07e005456359ca488fdd3a94bbab",
"body": "## Purpose ##\n* Ensure that code in LLM Compressor is compliant with the\n`torch.accelerator` api, which is superior to `torch.cuda` due to it\nbeing compatible with the XPU interface\n\n## Example ##\n```\n$ make quality\nRunning python quality checks\nruff check src tests examples setup.py;\nAll checks pas\n[…]\n setup.py; \npython tools/lint_cuda.py src tests examples setup.py --fix\nFixed 1 issue(s) in examples/autoround/ddp/ddp_qwen3_example.py\n```\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[XPU] Add `torch.cuda` linter (#2776)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-06T17:52:09Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "4c1a328855026b00dde8009b34d222393470fe8c",
"body": "…rom yaml (#2822)\n\n## Purpose ##\n* Remove deprecation warnings which are triggered when using a recipe\nfrom yaml\n\n## Changes ##\n* Instead of importing packages and then filtering (which leads to\ndeprecated packages being imported), instead filter packages then import\n\n## Testing ##\n* No more deprecation warnings in tests for modifiers\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Remove deprecation warnings which are triggered when using a recipe f…",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-06T17:51:23Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "6846bae790250937853c29113f6bf2ec406f8e2f",
"body": "## Purpose ##\n* Support optimal loading for a larger variety of MoEs, including those\nfrom transformers\n[v5.13.0](https://github.com/huggingface/transformers/releases/tag/v5.13.0)\n* Expand testing for a larger variety of MoEs\n* Move to an MoE import pattern which supports backwards compatibility &\ns\n[…]\nd mappings\n\n## Testing ##\n* Passes CI (which uses older transformers version)\n* Passes locally with newer (unreleased) transformers version\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[MoE] Support more MoE model architectures (#2847)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-06T16:50:41Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "2e7456ae9362b6a7d14b3e9889dcdb5c2930759d",
"body": "Reflects https://github.com/vllm-project/compressed-tensors/pull/702\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Docs] Update W8A16 docs (#2700)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-03T16:56:46Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "973c9c539a84dd9efaf74e115ede5ca419704c18",
"body": "## Purpose ##\n* Reduce the verbosity and excess usage of coderabbit\n * Save free-tier usage for more important uses\n * Avoid clogging PR discussions\n\n## Changes ##\n* Turn off automatic summaries\n * @fynnsu and others have indicated that these are not useful\n * You can still trigger these manuall\n[…]\nat features.\n\n<!-- end of auto-generated comment: release notes by coderabbit.ai -->\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Fynn Schmitt-Ulms <fynnsu@outlook.com>",
"is_bot": false,
"headline": "[Agents] Configure coderabbit (#2809)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-01T22:40:09Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "c9e4b5dce0d674755e42582f3d4f1d9a28f3a1d4",
"body": "## Background ##\nAs of #2498, reindexing is no longer required\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>\nCo-authored-by: CodeRabbit <noreply@coderabbit.ai>",
"is_bot": false,
"headline": "[model_free_ptq] Deprecate `reindex_fused_weights` (#2737)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-07-01T22:20:13Z",
"body_truncated": false,
"is_coding_agent": true
},
{
"oid": "49e51709818b4437a1eeaccd39096402c34680ef",
"body": "SUMMARY:\n\"Migrate all torch.cuda.* call sites in src/ to the unified\ntorch.accelerator API.\"\n\n\nRef PR:\nhttps://github.com/vllm-project/llm-compressor/pull/2611\n\n---------\n\nSigned-off-by: Suluner <suluner2011@gmail.com>",
"is_bot": false,
"headline": "Replace torch.cuda with torch.accelerator in main code path. (#2884)",
"author_name": "Ting Hu",
"author_login": "suluner",
"committed_at": "2026-07-01T16:09:41Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "e48fec881f1394b18fa56969d555c79b9278be8e",
"body": "…ified defaults (#2875)\n\n## What this PR does\n\nAdds a Claude Code skill (`/fp8`) that generates a working FP8\nquantization example script for any HuggingFace model and saves a\ncompressed-tensors checkpoint. When a user asks to quantize a model to\nFP8, the skill gathers the necessary inputs, selects \n[…]\n.com/claude-code)\n\n---------\n\nSigned-off-by: Dipika Sikka <ds3822@columbia.edu>\nCo-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Agents] Update fp8 skill with MoE/VLM architecture guidance and clar…",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-07-01T00:08:26Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "86f4cf1a9ca933bf52d830a4ac34a2e21fd13848",
"body": "## Purpose ##\n* Add a tool for creating tiny, finetuned test models which have the\nsame model architecture as their parents, but with fewer parameters\n\n## Usage ##\n\n```\n/create-tiny-model Qwen/Qwen3-30B-A3B\n```\n\n## Steps to creating a tiny model ##\n\nWhen this skill is invoked, the following steps wi\n[…]\naude will attempt to use `huggingface-cli`, despite being explicitly\ntold to use `hf`\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Dipika Sikka <dipikasikka1@gmail.com>",
"is_bot": false,
"headline": "[Claude] `create-tiny-model` skill for creating tiny test models (#2753)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-30T19:42:47Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "da6e4c2a9dc10908e5fcac89a8ec851dbd7fe80a",
"body": "SUMMARY:\n\nAdded a note in the project README \"What's New\" section about the new\nREAPPruningModifier\n\n---------\n\nSigned-off-by: Ryan Fernandes <ryan@fernandes.us>",
"is_bot": false,
"headline": "REAP added to What's New in README (#2881)",
"author_name": "Ryan Fernandes",
"author_login": "Ryfernandes",
"committed_at": "2026-06-30T16:57:52Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "82d697a6c9e65ea036fba42af81546cf7013cd4a",
"body": "# REAP Expert Pruning for MoE Models\n\nThis PR implements REAP (Router-weighted Expert Activation Pruning) for\nstructurally compressing Mixture of Experts (MoE) models.\n\n## Overview\n\nREAP reduces the VRAM footprint of MoE models by structurally removing\nindividual experts in their MoE layers based on\n[…]\n.noreply.github.com>\nSigned-off-by: Ryan Fernandes <ryan@fernandes.us>\nCo-authored-by: Eldar Kurtic <8884008+eldarkurtic@users.noreply.github.com>\nCo-authored-by: Dipika Sikka <dipikasikka1@gmail.com>",
"is_bot": false,
"headline": "REAP Pruning Modifier (#2864)",
"author_name": "Ryan Fernandes",
"author_login": "Ryfernandes",
"committed_at": "2026-06-30T15:44:46Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "930e5b8d0948c3734c2aa1dfd4165bc6045187c5",
"body": "… saving (#2868)\n\n## Purpose ##\n* Support saving models which take more than 10 minutes to save\n* Longer save times can be exacerbated by very large models and disk/cpu\noffloading\n\n## Changes ##\n* Add `suspend_distributed_timeout`\n\n## Testing ##\n* Add `test_suspend_distributed_timeout`\n* Now able to save models which take +20min to save\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Distributed] Add `suspend_distributed_timeout` to enable large model…",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-29T01:19:37Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "e3a4ac1fdfbaf5803052352abe9c834d3965ca42",
"body": "## Summary\n- Add `Qwen2_5_VLForConditionalGeneration` to the AWQ mapping registry\nwith `default_mappings`\n- Add four models to the SmoothQuant registry that were already present\nin AWQ: `Qwen2_5_VLForConditionalGeneration`,\n`Qwen2_5OmniThinkerForConditionalGeneration`, `SeedOssForCausalLM`,\n`Ernie4_\n[…]\n@gmail.com>\nCo-authored-by: OpenClaw Agent <agent@openclaw.local>\nCo-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "fix: add missing model mappings for AWQ and SmoothQuant (#2727)",
"author_name": "Asad Shahid",
"author_login": "AsadShahid04",
"committed_at": "2026-06-26T21:53:23Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "3707778241acf08eeb2f6aee7f4f50b02b2b45ca",
"body": "## Summary\n1) I noticed that even when DDP and single gpu models were exactly the\nsame, inference was giving different results because they were\ndispatched differently (auto offload vs on-device) aligning this makes\nthings a ton more consistent.\n2) made all the tests do activation calibration otherw\n[…]\non_ddp.py\n\n🤖 Generated with [Claude Code](https://claude.com/claude-code)\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>",
"is_bot": false,
"headline": "fix: DDP smoke test improvements and device_map mismatch fix (#2863)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-25T14:10:33Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "0dcfa37fd77f31707ba09da98f5c283c2d2c95e4",
"body": "## Summary\nthis sparsity stuff makes gptq data dependent and to my knowledge its\ncompletely unused\n\n## Test plan\n- [x] `pytest tests/llmcompressor/modifiers/gptq/ -v` — 53 passed, 1\nskipped\n\n🤖 Generated with [Claude Code](https://claude.com/claude-code)\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>",
"is_bot": false,
"headline": "Remove sparsity-preserving logic from GPTQ quantization (#2860)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-24T18:30:08Z",
"body_truncated": false,
"is_coding_agent": true
},
{
"oid": "31f223bd8aedc99e15920f488b196849b0c60198",
"body": "Summary\nBoth ranks are trying to clean up the temp dirs causing a race condition\n\nfail:\n\nhttps://github.com/neuralmagic/llm-compressor-testing/actions/runs/27996125653/job/82858691361\n\n\nTEST PLAN:\n\nhttps://github.com/neuralmagic/llm-compressor-testing/actions/runs/28053514637\n\n(no race conditions)\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>",
"is_bot": false,
"headline": "fix ddp smoke race conditions (#2857)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-24T15:00:15Z",
"body_truncated": false,
"is_coding_agent": true
},
{
"oid": "99df48a08dff500145776abb490ea0ee9d36fe17",
"body": "- Fix FP8 Examples link in docs\n\nSigned-off-by: Dipika Sikka <dipikasikka1@gmail.com>",
"is_bot": false,
"headline": "Update link to FP8 quantization documentation (#2854)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-06-24T13:23:55Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "c8975b8f13879e79a52c0af184958b3ffd5f7d2c",
"body": "…2855)\n\n## Summary\nwhen we [updated the\nlifecycle](https://github.com/vllm-project/llm-compressor/pull/2784)\nautoround was broken because apply autoround expects a decoder_layer\nspecifically and we added a filter to only pass the quantized linear\nlayers instead (like all the other modifiers) removin\n[…]\nests/llmcompressor/modifiers/autoround/test_base.py`\n\nPR made with claude\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>",
"is_bot": false,
"headline": "fix: remove is_module_quantized filter that made AutoRound a no-op (#…",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-23T20:39:18Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "5bebc2baf4e97d3e0b7922139d7d0f899c87f250",
"body": "SUMMARY:\n\"Add Mellum2 FP8 and NVFP4 Example\"\n\n---------\n\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Add Mellum2 Example (#2832)",
"author_name": "Krishna Teja Chitty-Venkata",
"author_login": "krishnateja95",
"committed_at": "2026-06-22T20:48:51Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "5c30810b5fb08277338584841040acc6f1f0f027",
"body": "…(#2785)\n\n## Purpose ##\n* Accelerate distributed calibration for `QuantizationModifier` by\nperforming weight calibration in parallel\n* The speedups from this change are minor: runtime is typically\ndominated by activation calibration, not weight calibration\n\n## Changes ##\n* Implement weight-parallel \n[…]\n- | - |\n| 1 | 8.69m | 14m |\n| 2 | 7.72m | 15m |\n| 3 | 8.09m | 16m |\n\nNon-linear speedups are due to fixed costs associated with calibration\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Distributed] Module parallel calibration for `QuantizationModifier` …",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-22T20:43:47Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e5d1ef1c44f2a2e2827c3aa5c9bc6cc443039571",
"body": "…types, e.g. diffusion_gemma) (#2845)\n\n## Summary\n\nBump the **release** upper bound on `transformers` from `<=5.10.1` to\n`<=5.12.1`.\n\nThe current release pin (`setup.py`) caps\n`transformers>=5.9.0,<=5.10.1`. That upper bound predates several recent\nHuggingFace model types — notably **`diffusion_gemm\n[…]\nr a different ceiling (or dropping the upper bound\nentirely, as some ecosystems do), happy to adjust.\n\nSigned-off-by: Jetha Chan <jethachan@gmail.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Bump tested transformers upper bound to 5.12.1 (unblocks newer model …",
"author_name": "Jetha Chan",
"author_login": "jethac",
"committed_at": "2026-06-22T19:46:43Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e72d639a3a743b215d0e3d9c5360c116e0fb1f35",
"body": "- Closes: https://github.com/vllm-project/llm-compressor/issues/2386\n\n## SUMMARY:\n- Add mappings for\nhttps://huggingface.co/stepfun-ai/Step-3.5-Flash/blob/main/modeling_step3p5.py\n- Add example script (untested) in `step3p5_example.py`\n\nThanks so much!\n\n## Architecture Notes\nStep-3.5-Flash:\n- **GQA \n[…]\nompressor/modifiers/transform/awq/test_dynamic_mappings.py`\n\n---------\n\nSigned-off-by: wanadzhar913 <adzhar.faiq@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Add AWQ model mappings for Step3p5 (#2770)",
"author_name": "Faiq Adzlan",
"author_login": "wanadzhar913",
"committed_at": "2026-06-22T18:24:07Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "ae83fdd1e6131390d540f2c1ece869073de71496",
"body": "…bjects (#2839)\n\n## Summary\n- Widen the `recipe` parameter type on `oneshot()` from `str | list[str]\n| None` to `RecipeInput | None`\n- Update docstring to reflect accepted input types\n\n`RecipeArguments.recipe` is left as `str | None` since\n`HfArgumentParser` needs primitive types.\n\nFixes #2837\n\n----\n[…]\nned-off-by: Sawyer Bowerman <sbowerma@redhat.com>\nCo-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "fix: widen oneshot() recipe type hint to accept Modifier and Recipe o…",
"author_name": "Sawyer Bowerman",
"author_login": "soyr-redhat",
"committed_at": "2026-06-22T18:01:51Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "e92fe38c98d73a32fbdf22d0e5ce34b4fd986468",
"body": "Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Bugfix] [MoE] Fix qwen3 loading (#2848)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-22T17:54:12Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "4017b8f4788aac43eac14e05c6d7a0ea44dd3d4e",
"body": "fixing CI issues with DDP smoketests\n\nfailed:\nhttps://github.com/neuralmagic/llm-compressor-testing/actions/runs/27731361640/job/82039025613\n\nAutoround has an issue so setting that to skip, unsure what the problem\nis as of now\n\n`2026-06-18T06:51:55.0888501Z [torchrun0]:[rank0]: FileNotFoundError:\n[E\n[…]\n://github.com/neuralmagic/llm-compressor-testing/actions/runs/27788891140\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>",
"is_bot": false,
"headline": "fix: DDP smoke test CI failures (#2840)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-22T17:35:13Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "b687eb821932643407e598203d0bb5f4481f7ac8",
"body": "Summary\n\nwhen any of the fused layers are missing (shared k,v in qkv) this will\nerror without this change, see the unit test\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>",
"is_bot": false,
"headline": "fix: skip None layers in fuse_weight_observers (#2841)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-22T16:51:25Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "eae3af52f3856cb88b7fd461e156854091a0cb07",
"body": "https://github.com/huggingface/transformers/blob/7bd2725e97c786a85180df0dea5441b290a9b1d6/src/transformers/models/auto/tokenization_auto.py#L707-L708\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Misc] Remove `use_fast` arg (#2849)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-22T16:46:02Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "90fb24b205f031db05a3cd4f8aa389192d307036",
"body": "## Background ##\nAs of the introduction of `load_offloaded_model` via\nhttps://github.com/vllm-project/llm-compressor/pull/2148 and\nhttps://github.com/vllm-project/llm-compressor/pull/2373, users are\nencouraged to load models as offloaded. This allows for distributed\ndispatches, disk offloading etc. \n[…]\nch the pipeline dispatch).\n\n## Purpose ##\n* Avoid excess device movement and OOM errors by not having `datafree`\nand `basic` pipelines dispatch models\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Remove all device movement from pipelines (#2846)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-22T16:43:04Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "10812a81500b87b4c522d38a69d293b954e58e1e",
"body": "Summary:\n\npreviously this was fixing a bug, the bugfix was moved\nhttps://github.com/vllm-project/compressed-tensors/pull/739 but this\nexample is still useful\n\n---------\n\nSigned-off-by: HDCharles <{\"message\":\"Not Found\",\"documentation_url\":\"https://docs.github.com/rest/users/emails#list-email-addresses-for-the-authenticated-user\",\"status\":\"404\"}>\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>",
"is_bot": false,
"headline": "gemma4 example (#2816)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-18T22:34:30Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "dc943cd99200de43cab0b532c6ac6aa3a03127d8",
"body": "…833)\n\n**SUMMARY**:\n\n#2677 removed the .weight suffix from split tensor names based on the\nassumption that safetensors tensor names always include `.weight`. That\nwas a valid fix for the case introduced in #2464: when the input tensor\nalready ended in .weight, the #2464 logic could produce invalid n\n[…]\nng validation so validation and quantization\noperate on the same tensor names.\n\n**TEST PLAN**:\nminor regression\n\n---------\n\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "[model_free] fix: skipped quantization on split fused moe experts (#2…",
"author_name": "LeonEricsson",
"author_login": "LeonEricsson",
"committed_at": "2026-06-18T17:12:47Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e16b807350578d397045d33a2f7e2a90df5c3b2d",
"body": "…#2769)\n\n## Summary\nI noticed these smoke tests for smoothquant and figured it would be\nuseful to have something similar/more comprehensive for all the\nmodifiers/techniques\n\n- Add multi-GPU DDP smoke tests covering all compression modifiers\n(SmoothQuant, RTN minmax/MSE/imatrix, CPU offload, disk off\n[…]\ntest_compression_ddp.py -m\nmulti_gpu -v\n\n(took about 8 minutes)\n \nsee CI\n\n🤖 Generated with [Claude Code](https://claude.com/claude-code)\n\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>",
"is_bot": false,
"headline": "[Tests] Add comprehensive DDP smoke tests for compression modifiers (…",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-17T18:42:47Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "09ac68642471bda683406abb52fb8cb1ea220556",
"body": "## summary\ntorch.compile support for MSE observer with chunked execution, opt-in\nvia `set_torch_compile(True)`. `Default off` (no behavior change).\n\n## benchmarks (A100 SXM 80GB / RunPod)\n\n### e2e \n\n(64 samples, max_seq=4096 for activation quantization)\n\n| | +activation (Qwen3-8B) | -activation (Qwe\n[…]\n.github.com>\nCo-authored-by: HDCharles <charlesdavidhernandez@gmail.com>\nCo-authored-by: Claude Opus 4.6 <noreply@anthropic.com>\nCo-authored-by: HDCharles <39544797+HDCharles@users.noreply.github.com>",
"is_bot": false,
"headline": "perf: make MSE observer compatible with torch.compile (#2384)",
"author_name": "Bias92",
"author_login": "Bias92",
"committed_at": "2026-06-17T17:52:49Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "a6aa5003ee934fcb81e9793558fbc1e65de15f77",
"body": "## Purpose ##\n* Solidify the interface for `on_sequential_epoch_end` by passing\nmodules directly\n* Reduce boilerplate for extracting `modules` from method kwargs\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Lifecycle] Add `modules` argument to `on_sequential_epoch_end` (#2784)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-17T17:39:02Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "435eed9984979bbc9ab1aa06e738ac1770a888b5",
"body": "…n (#2718)\n\n## Summary\nThis PR improves SmoothQuant mapping inference for **Qwen3.5\nhybrid-attention models (both dense and MoE variants)** and aligns the\nimplementation structure more closely with the existing dynamic mapping\nflow used by AWQ.\n\nThe main issue is that Qwen3.5 only exposes\n`self_attn\n[…]\nd-off-by: zhangxin81 <115389973+zhangxin81@users.noreply.github.com>\nSigned-off-by: zhangxin81 <zxinxinxinxinxin@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "build SmoothQuant mapping dynamically for Qwen3.5 MoE hybrid attentio…",
"author_name": "zhangxin.zzzzz",
"author_login": "zhangxin81",
"committed_at": "2026-06-17T16:41:13Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "b2a5795a2d72cca97e8cc67b4c37ede21c3ede62",
"body": "…2798)\n\n## Summary\nWhen a model with tied word embeddings is saved while offloaded, the\noffload conversion (`to_accelerate`) splits `embed_tokens` and `lm_head`\ninto separate parameters. This defeats transformers' save-time\nde-duplication, so a redundant, bit-identical `lm_head.weight` gets\nwritten.\n[…]\nti@cohere.com>\nSigned-off-by: Brian Dellabetta <bdellabe@redhat.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>\nCo-authored-by: Brian Dellabetta <bdellabe@redhat.com>",
"is_bot": false,
"headline": "fix: re-tie word embeddings before save to avoid duplicate lm_head (#…",
"author_name": "EdalatiAli",
"author_login": "EdalatiAli",
"committed_at": "2026-06-16T19:12:13Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "4ec119d3460c34cb4b2eed22c64cb926066cb209",
"body": "## Purpose\n\nAdds `examples/quantization_embedding/` — an example showing how to\nquantize a model's **input embedding table** to weight-only `intN`\n(WNA16) with a data-free `QuantizationModifier`. vLLM loads these\ncheckpoints (embedding-quant support added in vllm-project/vllm#44340)\nand runs a fused\n[…]\n-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Add input-embedding quantization example (#2830)",
"author_name": "Karthik Kothuri",
"author_login": "KKothuri",
"committed_at": "2026-06-15T21:58:52Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "715e1166c9263124548da9d2e307fdb4c7f17eab",
"body": "## Purpose ##\n* Remove confusing and boiler-plate modifier code for handling\ncalibration events\n* Rename `calibration_epoch_start/end` to `calibration_start/end` for\nclarity, as requested by @brian-dellabetta\n\n## Changes ##\n* Add calibration lifecycle event hooks as optional methods that\nmodifiers c\n[…]\now handled by the\n`Modifier` base class\n* Updated documentation \n\n## Testing ##\n* All tests in `tests/llmcompressor/modifiers/` folder pass\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Lifecycle] Make calibration events first-class citizens (#2783)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-15T21:41:52Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "e1a69477143c35f1e50d4d2bdf49f68e79cd7402",
"body": "Summary:\n\nobservers need stats of their fused observers to calculate global_scale\nproperly.\n\nwe run all observers in a chunk at the same time so normally this isn't\nan issue.\n\nif sequential targets is set to linear or something else small, the\nfused observers will not be in the current chunk and so \n[…]\n't matter but in the situation where the\nsequential targets is really small, it will solve the problem.\n\nTest Plan:\nsee unit test\n\n---------\n\nSigned-off-by: HDCharles <charlesdavidhernandez@gmail.com>",
"is_bot": false,
"headline": "fix bug fused obs + sequential targets (#2732)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-15T20:39:31Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "bfd4b40a1172cef01e36f41793e201f3b3237f77",
"body": "## Purpose ##\n* Remove warning for gold-standard and expected behavior\n* For context, this warning was added when passing `output_dir` to\n`oneshot` or `train` was the default flow\n * Now, we discourage users from passing `output_dir`\n* It's been almost two years since we made this change, it's reas\n[…]\nve warning\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>\nCo-authored-by: CodeRabbit <noreply@coderabbit.ai>",
"is_bot": false,
"headline": "[Misc] Remove oneshot `output_dir` warning (#2782)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-15T19:08:54Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "09f9a944bac1435f021d44828525360365d430f0",
"body": "Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "[QuIP] Change default to only use input (v) rotations (#2815)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-15T18:44:05Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "13086b301505abc37576801157746d7f33728fdf",
"body": "## Purpose ##\n* Drastically speed up validation step for model_free_ptq\n\n# Note! #\nThis branch depends on\nhttps://github.com/vllm-project/compressed-tensors/pull/726. It's likely\nthat a release will be cut after\nhttps://github.com/vllm-project/llm-compressor/pull/2647 lands, so it's\nbest to let the other PR land first, release, then merge this one in.\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[model_free_ptq] Perform validation on meta device (#2796)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-15T16:32:14Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "7325301c5d4d34a211c8e59d4e749a62c385edf5",
"body": "```bash\nvllm ({'pretrained': 'Qwen3-8B-W4A16-G128-AutoRound-Ultrachat200k/', 'tensor_parallel_size': 1, 'max_model_len': 8192, 'max_num_batched_tokens': 32768, 'max_num_seqs': 128, 'add_bos_token': True, 'gpu_memory_utilization': 0.8, 'dtype': 'bfloat16', 'max_gen_toks': 2048, 'enable_prefix_caching\n[…]\nd\"\n\n---------\n\nSigned-off-by: yiliu30 <yi4.liu@intel.com>\nSigned-off-by: Yi Liu <yi4.liu@intel.com>\nCo-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>",
"is_bot": false,
"headline": "[Autoround] add custom dataset support (#2824)",
"author_name": "Yi Liu",
"author_login": "yiliu30",
"committed_at": "2026-06-15T16:07:49Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "20563394f9e1f6d64f59ab5f727b078cd4cd3eeb",
"body": "## Purpose ##\n* Remove unnecessary autowrapping for if condition which contain `:=` in\nthe statement but not the condition\n\nStill wrapped:\n```python3\nif (a := b):\n ...\n```\n\nNo longer wrapped:\n```python3\nif True:\n a := b\n```\n\n## Testing ##\n* Walrus operator tests in `test_autowrapper` still pass\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Bugfix] [Autowrapper] Only check walrus in if condition (#2819)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-15T15:36:50Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "d54355f5df94b4c2eb38ab89084e244b60a42612",
"body": "Summary\n- Mention Transformers v5 supoprt\n- Link DiffusionGemma models",
"is_bot": false,
"headline": "Update What's New (#2831)",
"author_name": "Dipika Sikka",
"author_login": "dsikka",
"committed_at": "2026-06-15T14:56:59Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "6d2a0908e45324b76e390bdec600143e85a075f8",
"body": "## Purpose ##\n* Fix calibration of all experts\n* Need to enter `moe_calibration_context` during oneshot. In the future,\nI'd like to add this to `calibration_forward_context` and deprecate the\n`moe_calibrate_all_experts` argument\n* I'm not sure why this error was never seen by any of the other NVFP4/\n[…]\ntware industry. I'm looking for a job in the software industry. I'm looking for a job in the\n==========================================\n```\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[MoE] [Bugfix] Add missing `moe_calibration_context` (#2827)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-12T16:59:22Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "a32b17d2fa8c2b4432576c785ed7eeb311bfeb45",
"body": "SUMMARY:\nUpdate dep versions for release.\n\n\nTEST PLAN:\nAll tests.\n\n---------\n\nSigned-off-by: Dan Huang <dahuang@redhat.com>",
"is_bot": false,
"headline": "Update dep versions for release (#2810)",
"author_name": "dhuangnm",
"author_login": "dhuangnm",
"committed_at": "2026-06-11T19:55:53Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "ccdd1e7ec1eb98d600fbef3f9afcc1da06874a2b",
"body": "Fixes #2696\n\n## Summary:\nAdds AWQ smoothing mappings for `Glm4MoeLiteForCausalLM` (GLM-4.7-Flash\nmodel).\n\n## Architecture Notes\nGLM-4.7-Flash:\n- **MLA attention** (`q_a_proj`, `kv_a_proj_with_mqa`, `q_b_proj`,\n`kv_b_proj`) instead of standard `q/k/v_proj`\n- **Mixed dense/MoE FFN**: `first_k_dense_re\n[…]\nully.\nVerified inference via vLLM produces coherent outputs.\n\nSigned-off-by: Priyadharshini <gpriyadharshini0205@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "Add AWQ Mappings for Glm4MoeLiteForCausalLM (#2768)",
"author_name": "Priyadharshini",
"author_login": "Priya95715",
"committed_at": "2026-06-11T14:02:50Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "52eec9a04f760ee993d52872fa3a5e0085b3add2",
"body": "## Purpose ##\n* Fix KV Cache tests when using `init_empty_weights` with\n`accelerate==1.6.0` and `transformers>=5.9.0`\n* Newer versions of transformers attach an attribute\n`_is_hf_initialized` to parameters. However,\n`accelerate::init_empty_weights` treats this as a class constructor\nparam. When it t\n[…]\nt_empty_weights` and instead loading on the meta device instead.\n\n## Changes ##\n* Remove all uses of `init_empty_weights`, instead load on meta device\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Bugfix] Fix KV cache tests with transformers v5 (#2823)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-10T21:48:18Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "539b03e41f0bd028acf424c1abf61e8633dfb19c",
"body": "## Summary\nThis PR introduces a new unified context manager `load_context` that\ncombines both `load_offloaded_model` and `load_quantizable_moe`\ncontexts, simplifying model loading code across the codebase.\n\n## Changes\n\n### Core Implementation\n- **Add `load_context()` in `src/llmcompressor/utils/dev.\n[…]\nm>\nCo-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>\nCo-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>\nCo-authored-by: CodeRabbit <noreply@coderabbit.ai>",
"is_bot": false,
"headline": "Add `load_context` unified context manager (#2811)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-10T20:49:57Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "a3efcbe0b35004b71046d6c769039c19619fa609",
"body": "## Purpose ##\n* Explain how the new MoE loading system works\n* Explain how to add support for MoE models which do not follow the\n`use_experts_implementation` convention\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Docs] Add MoE + Transformers V5 docs (#2820)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-10T20:24:38Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "650d782694b80bc1639331dbeeb1a2fd6f2290ab",
"body": "Summary:\n\nimatrix creates its hook differently from hooks_mixin so it wasn'te\nrespecting this\n\nkind of hacky but we're going to rewrite this anyway when we get rid of\nimatrixgatherer\n\nTEST PLAN:\npytest\n/home/HDCharles/repos/llm-compressor/tests/llmcompressor/observers -k\ntest_hook\n\n---------\n\nSigned\n[…]\nlesdavidhernandez@gmail.com>\nSigned-off-by: HDCharles <39544797+HDCharles@users.noreply.github.com>\nCo-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>",
"is_bot": false,
"headline": "fix imatrix + awq (#2814)",
"author_name": "HDCharles",
"author_login": "HDCharles",
"committed_at": "2026-06-09T19:48:51Z",
"body_truncated": true,
"is_coding_agent": true
},
{
"oid": "a71ae2344bc6759b7127b297334e24fc3a0ae5aa",
"body": "# Transformers v5 #\n* Resolves #2159\n\n## Examples and Model Loading ##\n\n* `dtype` now defaults to `\"auto\"`, so we can remove this explicit\nargument to reduce confusion\n* `from_pretrained` no longer supports `use_auth_token`\n* Model structures differ between v4 and v5. This means that our example\nreg\n[…]\nfor testing that regexes still match after v5\nupgrade\nhttps://github.com/vllm-project/llm-compressor/pull/new/kylesayrs/example_regex_tests\n\n---------\n\nSigned-off-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "Transformers v5 (#2647)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-09T18:29:47Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "4818655c5bac230f2ee35957427457dd939c0251",
"body": "## Summary\n\n- Removes the stale `# TODO: this does not serialize / will be\nincorrectly written` comment on `GPTQModifier.actorder`\n\nThe TODO was added in #1815 when serialization behavior of the\n`Sentinel(\"static\")` default was uncertain. It is no longer applicable:\n\n- `test_serialize_actorder` conf\n[…]\nd-tensors==0.16.0`).\n\n## Test plan\n\n- No behavior change — comment-only removal\n- Existing `test_serialize_actorder` covers serialization correctness\n\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "cleanup: remove stale actorder serialization TODO (#2807)",
"author_name": "Sawyer Bowerman",
"author_login": "soyr-redhat",
"committed_at": "2026-06-09T14:56:27Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "003ccbcfbc3b1a8b7216f7c8d9eaa47e043dca8f",
"body": "…) (#2589)\n\n## Summary\nCloses #2551\nSimplifies the `splits` interface in\n[get_processed_dataset](cci:1://file:///data/arpit/projects/llm-compressor/src/llmcompressor/datasets/utils.py:28:0-94:32)\nby removing\nmulti-split dict handling in favour of a plain string argument.\n\n\n### Examples & Tests\n- Upd\n[…]\ned-off-by: Arpit <arpit@example.com>\nCo-authored-by: Arpit <arpit@example.com>\nCo-authored-by: HDCharles <39544797+HDCharles@users.noreply.github.com>\nCo-authored-by: Kyle Sayers <kylesayrs@gmail.com>",
"is_bot": false,
"headline": "[Refactor] Refactor splits to only use the \"calibration\" split (#2551…",
"author_name": "Arpit Khandelwal",
"author_login": "arpitkh101",
"committed_at": "2026-06-08T01:53:35Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "6a5c9bde488ddd4026377b336263994e48d9eb2d",
"body": "SUMMARY:\nMinor textual and semantic fixes across docs and setup:\n- `docs/index.md`: fixed missing word \"to\" in sparse compression\ndeprecation warning\n- `docs/guides/big_models_and_distributed/model_loading.md`: fixed\nspelling of \"auxiliary\"\n- `setup.py`: fixed malformed f-string literal in version logging\n\nTEST PLAN:\nText-only changes in docs. The setup.py fix corrects a debug log string\nwith no impact on build behavior.",
"is_bot": false,
"headline": "fix: semantic and textual linting updates (#2795)",
"author_name": "Sawyer Bowerman",
"author_login": "soyr-redhat",
"committed_at": "2026-06-07T14:34:42Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "4e6c34c3194fcdb2e17dae090eed44d97d5397eb",
"body": "Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>\nSigned-off-by: Dipika Sikka <dipikasikka1@gmail.com>\nCo-authored-by: Dipika Sikka <dipikasikka1@gmail.com>",
"is_bot": false,
"headline": "[Examples] [model_free_ptq] Nemotron Ultra Example (#2803)",
"author_name": "Kyle Sayers",
"author_login": "kylesayrs",
"committed_at": "2026-06-05T13:49:40Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "e183fed8527c68ec1a27835fc486fd621a248d0f",
"body": "SUMMARY:\nThe `[tool.mypy]` config in `pyproject.toml` had `files =\n\"src/guidellm\"`, which appears to be an artifact from guidellm. Updated\nto `files = \"src/llmcompressor\"` so mypy checks the correct source tree.\n\nTEST PLAN:\nVerified that `src/llmcompressor` exists and `src/guidellm` does not.\nThis is a config-only fix with no runtime impact.",
"is_bot": false,
"headline": "fix: correct mypy files path from guidellm to llmcompressor (#2794)",
"author_name": "Sawyer Bowerman",
"author_login": "soyr-redhat",
"committed_at": "2026-06-04T19:59:14Z",
"body_truncated": false,
"is_coding_agent": false
},
{
"oid": "29461a1959fa8f4cbd8fe3f8cae991e4e6588291",
"body": "## SUMMARY\n\nPart of #1927. Modernizes type hints in `src/llmcompressor/utils/` to\nuse PEP 604 union syntax (`X | Y`) and PEP 585 built-in generics\n(`list`, `dict`).\n\nTwo files touched:\n- `src/llmcompressor/utils/metric_logging.py`\n- `src/llmcompressor/utils/pytorch/module.py`\n\nAlso moves `Iterable` \n[…]\n no other\nmodule imports the removed `typing` names from these files.\n\nSigned-off-by: u7k4rs6 <utkarshbahuguna10@gmail.com>\nCo-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>",
"is_bot": false,
"headline": "refactor: modernize type hints in utils/ module (part of #1927) (#2777)",
"author_name": "Utkarsh Bahuguna",
"author_login": "u7k4rs6",
"committed_at": "2026-06-04T19:46:23Z",
"body_truncated": true,
"is_coding_agent": false
},
{
"oid": "5112c8bfdcc8abd4a1816a31139b33c8f90fccfc",
"body": "…2779)\n\n## SUMMARY\n\nPart of #1927. Second PR in the umbrella (after #2777). Modernizes type\nhints in `src/llmcompressor/modeling/` to use PEP 604 union syntax (`X |\nY`) and PEP 585 built-in generics (`list`, `tuple`).\n\nTwo files touched:\n- `src/llmcompressor/modeling/gpt_oss.py`\n- `src/llmcompressor\n[…]\nquires=\">=3.10\"`.\n- No downstream importer relied on the removed `typing.List` /\n`typing.Optional` / `typing.Tuple` re-exports from these modules.\n\nSigned-off-by: u7k4rs6 <utkarshbahuguna10@gmail.com>",
"is_bot": false,
"headline": "refactor: modernize type hints in modeling/ module (part of #1927) (#…",
"author_name": "Utkarsh Bahuguna",
"author_login": "u7k4rs6",
"committed_at": "2026-06-04T14:30:46Z",
"body_truncated": true,
"is_coding_agent": false
}
],
"releases_count": 27,
"commits_last_year": 654,
"latest_release_at": "2026-06-26T00:26:26Z",
"latest_release_tag": "0.7.1.3",
"releases_from_tags": false,
"days_since_last_push": 0,
"active_weeks_last_year": 50,
"days_since_latest_release": 29,
"mean_days_between_releases": 14.7
},
"community": {
"has_readme": true,
"has_license": true,
"has_description": true,
"has_contributing": true,
"health_percentage": 87,
"has_issue_template": false,
"has_code_of_conduct": true,
"has_pull_request_template": true
},
"ecosystem": {
"packages": [
{
"name": "llmcompressor",
"exists": true,
"license": "Apache 2.0",
"keywords": [
"llmcompressor",
"llms",
"large language models",
"transformers",
"pytorch",
"huggingface",
"compressors",
"compression",
"quantization",
"pruning",
"sparsity",
"optimization",
"model optimization",
"model compression",
"Development Status :: 5 - Production/Stable",
"Intended Audience :: Developers",
"Intended Audience :: Education",
"Intended Audience :: Information Technology",
"Intended Audience :: Science/Research",
"License :: OSI Approved :: Apache Software License",
"Operating System :: POSIX :: Linux",
"Programming Language :: Python :: 3",
"Programming Language :: Python :: 3 :: Only",
"Topic :: Scientific/Engineering",
"Topic :: Scientific/Engineering :: Artificial Intelligence",
"Topic :: Scientific/Engineering :: Mathematics",
"Topic :: Software Development",
"Topic :: Software Development :: Libraries :: Python Modules"
],
"ecosystem": "pypi",
"matches_repo": true,
"registry_url": "https://pypi.org/project/llmcompressor/",
"is_deprecated": false,
"latest_version": "0.12.0",
"repository_url": "https://github.com/vllm-project/llm-compressor",
"versions_count": 77,
"total_downloads": null,
"dependents_count": null,
"deprecation_note": null,
"maintainers_count": null,
"monthly_downloads": 190161,
"first_published_at": "2024-08-08T15:26:18.365386Z",
"latest_published_at": "2026-06-15T12:52:07.319596Z",
"latest_version_yanked": null,
"days_since_latest_publish": 39
}
]
},
"popularity": {
"forks": 586,
"stars": 3580,
"watchers": 29,
"fork_history": {
"days": [
{
"date": "2024-06-24",
"count": 1
},
{
"date": "2024-07-15",
"count": 1
},
{
"date": "2024-07-16",
"count": 1
},
{
"date": "2024-07-22",
"count": 1
},
{
"date": "2024-07-24",
"count": 1
},
{
"date": "2024-07-31",
"count": 1
},
{
"date": "2024-08-02",
"count": 1
},
{
"date": "2024-08-08",
"count": 1
},
{
"date": "2024-08-12",
"count": 1
},
{
"date": "2024-08-15",
"count": 1
},
{
"date": "2024-08-16",
"count": 2
},
{
"date": "2024-08-18",
"count": 2
},
{
"date": "2024-08-19",
"count": 3
},
{
"date": "2024-08-23",
"count": 1
},
{
"date": "2024-08-26",
"count": 2
},
{
"date": "2024-08-29",
"count": 1
},
{
"date": "2024-08-30",
"count": 1
},
{
"date": "2024-09-02",
"count": 1
},
{
"date": "2024-09-05",
"count": 2
},
{
"date": "2024-09-16",
"count": 1
},
{
"date": "2024-09-17",
"count": 2
},
{
"date": "2024-09-20",
"count": 1
},
{
"date": "2024-09-21",
"count": 1
},
{
"date": "2024-09-25",
"count": 2
},
{
"date": "2024-09-26",
"count": 4
},
{
"date": "2024-10-04",
"count": 1
},
{
"date": "2024-10-05",
"count": 2
},
{
"date": "2024-10-11",
"count": 1
},
{
"date": "2024-10-12",
"count": 1
},
{
"date": "2024-10-15",
"count": 2
},
{
"date": "2024-10-23",
"count": 1
},
{
"date": "2024-10-28",
"count": 1
},
{
"date": "2024-11-01",
"count": 1
},
{
"date": "2024-11-05",
"count": 2
},
{
"date": "2024-11-13",
"count": 1
},
{
"date": "2024-11-15",
"count": 1
},
{
"date": "2024-11-23",
"count": 1
},
{
"date": "2024-11-24",
"count": 1
},
{
"date": "2024-11-26",
"count": 1
},
{
"date": "2024-11-27",
"count": 1
},
{
"date": "2024-11-28",
"count": 1
},
{
"date": "2024-12-11",
"count": 3
},
{
"date": "2024-12-26",
"count": 1
},
{
"date": "2024-12-27",
"count": 1
},
{
"date": "2024-12-31",
"count": 2
},
{
"date": "2025-01-07",
"count": 1
},
{
"date": "2025-01-10",
"count": 1
},
{
"date": "2025-01-26",
"count": 1
},
{
"date": "2025-01-29",
"count": 1
},
{
"date": "2025-02-04",
"count": 1
},
{
"date": "2025-02-05",
"count": 2
},
{
"date": "2025-02-06",
"count": 1
},
{
"date": "2025-02-12",
"count": 2
},
{
"date": "2025-02-15",
"count": 2
},
{
"date": "2025-02-19",
"count": 2
},
{
"date": "2025-02-22",
"count": 1
},
{
"date": "2025-02-23",
"count": 1
},
{
"date": "2025-02-24",
"count": 1
},
{
"date": "2025-02-26",
"count": 1
},
{
"date": "2025-02-27",
"count": 2
},
{
"date": "2025-02-28",
"count": 1
},
{
"date": "2025-03-02",
"count": 1
},
{
"date": "2025-03-06",
"count": 1
},
{
"date": "2025-03-08",
"count": 1
},
{
"date": "2025-03-12",
"count": 2
},
{
"date": "2025-03-13",
"count": 1
},
{
"date": "2025-03-14",
"count": 1
},
{
"date": "2025-03-17",
"count": 1
},
{
"date": "2025-03-18",
"count": 1
},
{
"date": "2025-03-21",
"count": 1
},
{
"date": "2025-03-23",
"count": 1
},
{
"date": "2025-03-24",
"count": 1
},
{
"date": "2025-03-25",
"count": 2
},
{
"date": "2025-04-01",
"count": 1
},
{
"date": "2025-04-02",
"count": 1
},
{
"date": "2025-04-04",
"count": 1
},
{
"date": "2025-04-06",
"count": 2
},
{
"date": "2025-04-07",
"count": 1
},
{
"date": "2025-04-08",
"count": 2
},
{
"date": "2025-04-09",
"count": 1
},
{
"date": "2025-04-15",
"count": 1
},
{
"date": "2025-04-16",
"count": 3
},
{
"date": "2025-04-20",
"count": 1
},
{
"date": "2025-04-23",
"count": 2
},
{
"date": "2025-04-26",
"count": 1
},
{
"date": "2025-04-29",
"count": 1
},
{
"date": "2025-04-30",
"count": 1
},
{
"date": "2025-05-02",
"count": 2
},
{
"date": "2025-05-07",
"count": 1
},
{
"date": "2025-05-14",
"count": 1
},
{
"date": "2025-05-15",
"count": 1
},
{
"date": "2025-05-20",
"count": 1
},
{
"date": "2025-05-21",
"count": 1
},
{
"date": "2025-05-22",
"count": 2
},
{
"date": "2025-05-23",
"count": 2
},
{
"date": "2025-05-26",
"count": 1
},
{
"date": "2025-05-29",
"count": 1
},
{
"date": "2025-05-31",
"count": 1
},
{
"date": "2025-06-03",
"count": 1
},
{
"date": "2025-06-04",
"count": 1
},
{
"date": "2025-06-05",
"count": 1
},
{
"date": "2025-06-06",
"count": 1
},
{
"date": "2025-06-09",
"count": 1
},
{
"date": "2025-06-11",
"count": 1
},
{
"date": "2025-06-12",
"count": 1
},
{
"date": "2025-06-13",
"count": 2
},
{
"date": "2025-06-14",
"count": 1
},
{
"date": "2025-06-18",
"count": 1
},
{
"date": "2025-06-19",
"count": 2
},
{
"date": "2025-06-21",
"count": 1
},
{
"date": "2025-06-22",
"count": 1
},
{
"date": "2025-06-23",
"count": 1
},
{
"date": "2025-06-24",
"count": 1
},
{
"date": "2025-06-25",
"count": 1
},
{
"date": "2025-06-26",
"count": 1
},
{
"date": "2025-06-27",
"count": 2
},
{
"date": "2025-06-30",
"count": 3
},
{
"date": "2025-07-01",
"count": 2
},
{
"date": "2025-07-02",
"count": 1
},
{
"date": "2025-07-03",
"count": 5
},
{
"date": "2025-07-04",
"count": 1
},
{
"date": "2025-07-07",
"count": 1
},
{
"date": "2025-07-09",
"count": 2
},
{
"date": "2025-07-15",
"count": 2
},
{
"date": "2025-07-18",
"count": 2
},
{
"date": "2025-07-19",
"count": 1
},
{
"date": "2025-07-20",
"count": 1
},
{
"date": "2025-07-22",
"count": 1
},
{
"date": "2025-07-23",
"count": 1
},
{
"date": "2025-07-24",
"count": 2
},
{
"date": "2025-07-30",
"count": 2
},
{
"date": "2025-08-05",
"count": 1
},
{
"date": "2025-08-06",
"count": 4
},
{
"date": "2025-08-07",
"count": 1
},
{
"date": "2025-08-08",
"count": 1
},
{
"date": "2025-08-11",
"count": 3
},
{
"date": "2025-08-12",
"count": 1
},
{
"date": "2025-08-13",
"count": 2
},
{
"date": "2025-08-15",
"count": 1
},
{
"date": "2025-08-16",
"count": 1
},
{
"date": "2025-08-17",
"count": 1
},
{
"date": "2025-08-18",
"count": 1
},
{
"date": "2025-08-20",
"count": 3
},
{
"date": "2025-08-21",
"count": 2
},
{
"date": "2025-08-22",
"count": 2
},
{
"date": "2025-08-26",
"count": 1
},
{
"date": "2025-08-29",
"count": 2
},
{
"date": "2025-08-30",
"count": 1
},
{
"date": "2025-08-31",
"count": 1
},
{
"date": "2025-09-02",
"count": 1
},
{
"date": "2025-09-03",
"count": 1
},
{
"date": "2025-09-07",
"count": 1
},
{
"date": "2025-09-08",
"count": 1
},
{
"date": "2025-09-09",
"count": 1
},
{
"date": "2025-09-11",
"count": 1
},
{
"date": "2025-09-12",
"count": 1
},
{
"date": "2025-09-15",
"count": 1
},
{
"date": "2025-09-16",
"count": 1
},
{
"date": "2025-09-17",
"count": 2
},
{
"date": "2025-09-18",
"count": 1
},
{
"date": "2025-09-19",
"count": 1
},
{
"date": "2025-09-22",
"count": 1
},
{
"date": "2025-09-23",
"count": 1
},
{
"date": "2025-09-24",
"count": 2
},
{
"date": "2025-09-26",
"count": 1
},
{
"date": "2025-09-27",
"count": 2
},
{
"date": "2025-09-28",
"count": 1
},
{
"date": "2025-09-29",
"count": 1
},
{
"date": "2025-09-30",
"count": 3
},
{
"date": "2025-10-01",
"count": 2
},
{
"date": "2025-10-04",
"count": 1
},
{
"date": "2025-10-07",
"count": 1
},
{
"date": "2025-10-08",
"count": 1
},
{
"date": "2025-10-09",
"count": 1
},
{
"date": "2025-10-10",
"count": 1
},
{
"date": "2025-10-14",
"count": 3
},
{
"date": "2025-10-15",
"count": 1
},
{
"date": "2025-10-16",
"count": 1
},
{
"date": "2025-10-17",
"count": 1
},
{
"date": "2025-10-18",
"count": 1
},
{
"date": "2025-10-20",
"count": 2
},
{
"date": "2025-10-21",
"count": 1
},
{
"date": "2025-10-23",
"count": 1
},
{
"date": "2025-10-25",
"count": 1
},
{
"date": "2025-10-28",
"count": 2
},
{
"date": "2025-10-30",
"count": 1
},
{
"date": "2025-10-31",
"count": 1
},
{
"date": "2025-11-01",
"count": 1
},
{
"date": "2025-11-03",
"count": 3
},
{
"date": "2025-11-04",
"count": 1
},
{
"date": "2025-11-06",
"count": 2
},
{
"date": "2025-11-07",
"count": 1
},
{
"date": "2025-11-09",
"count": 2
},
{
"date": "2025-11-10",
"count": 2
},
{
"date": "2025-11-11",
"count": 1
},
{
"date": "2025-11-12",
"count": 2
},
{
"date": "2025-11-13",
"count": 1
},
{
"date": "2025-11-14",
"count": 1
},
{
"date": "2025-11-15",
"count": 1
},
{
"date": "2025-11-17",
"count": 1
},
{
"date": "2025-11-20",
"count": 1
},
{
"date": "2025-11-21",
"count": 3
},
{
"date": "2025-11-26",
"count": 1
},
{
"date": "2025-11-27",
"count": 1
},
{
"date": "2025-11-29",
"count": 1
},
{
"date": "2025-12-01",
"count": 1
},
{
"date": "2025-12-02",
"count": 2
},
{
"date": "2025-12-03",
"count": 2
},
{
"date": "2025-12-04",
"count": 2
},
{
"date": "2025-12-05",
"count": 3
},
{
"date": "2025-12-08",
"count": 4
},
{
"date": "2025-12-09",
"count": 2
},
{
"date": "2025-12-11",
"count": 1
},
{
"date": "2025-12-12",
"count": 2
},
{
"date": "2025-12-13",
"count": 1
},
{
"date": "2025-12-16",
"count": 3
},
{
"date": "2025-12-17",
"count": 2
},
{
"date": "2025-12-18",
"count": 3
},
{
"date": "2025-12-19",
"count": 3
},
{
"date": "2025-12-20",
"count": 1
},
{
"date": "2025-12-21",
"count": 2
},
{
"date": "2025-12-22",
"count": 2
},
{
"date": "2025-12-23",
"count": 1
},
{
"date": "2025-12-24",
"count": 1
},
{
"date": "2025-12-25",
"count": 1
},
{
"date": "2025-12-26",
"count": 2
},
{
"date": "2025-12-28",
"count": 4
},
{
"date": "2025-12-29",
"count": 3
},
{
"date": "2025-12-30",
"count": 2
},
{
"date": "2026-01-02",
"count": 1
},
{
"date": "2026-01-03",
"count": 1
},
{
"date": "2026-01-04",
"count": 1
},
{
"date": "2026-01-05",
"count": 1
},
{
"date": "2026-01-08",
"count": 3
},
{
"date": "2026-01-09",
"count": 1
},
{
"date": "2026-01-12",
"count": 4
},
{
"date": "2026-01-13",
"count": 2
},
{
"date": "2026-01-15",
"count": 2
},
{
"date": "2026-01-19",
"count": 4
},
{
"date": "2026-01-22",
"count": 1
},
{
"date": "2026-01-26",
"count": 2
},
{
"date": "2026-01-27",
"count": 2
},
{
"date": "2026-01-28",
"count": 1
},
{
"date": "2026-01-29",
"count": 4
},
{
"date": "2026-01-30",
"count": 2
},
{
"date": "2026-01-31",
"count": 2
},
{
"date": "2026-02-02",
"count": 1
},
{
"date": "2026-02-03",
"count": 1
},
{
"date": "2026-02-04",
"count": 3
},
{
"date": "2026-02-10",
"count": 2
},
{
"date": "2026-02-12",
"count": 2
},
{
"date": "2026-02-15",
"count": 2
},
{
"date": "2026-02-16",
"count": 3
},
{
"date": "2026-02-18",
"count": 4
},
{
"date": "2026-02-20",
"count": 1
},
{
"date": "2026-02-21",
"count": 2
},
{
"date": "2026-02-23",
"count": 1
},
{
"date": "2026-02-25",
"count": 3
},
{
"date": "2026-02-26",
"count": 3
},
{
"date": "2026-02-28",
"count": 1
},
{
"date": "2026-03-02",
"count": 2
},
{
"date": "2026-03-03",
"count": 2
},
{
"date": "2026-03-04",
"count": 3
},
{
"date": "2026-03-05",
"count": 1
},
{
"date": "2026-03-06",
"count": 1
},
{
"date": "2026-03-07",
"count": 1
},
{
"date": "2026-03-08",
"count": 1
},
{
"date": "2026-03-09",
"count": 6
},
{
"date": "2026-03-11",
"count": 2
},
{
"date": "2026-03-12",
"count": 2
},
{
"date": "2026-03-13",
"count": 1
},
{
"date": "2026-03-14",
"count": 1
},
{
"date": "2026-03-15",
"count": 1
},
{
"date": "2026-03-16",
"count": 1
},
{
"date": "2026-03-17",
"count": 1
},
{
"date": "2026-03-18",
"count": 2
},
{
"date": "2026-03-20",
"count": 2
},
{
"date": "2026-03-23",
"count": 4
},
{
"date": "2026-03-24",
"count": 2
},
{
"date": "2026-03-25",
"count": 1
},
{
"date": "2026-03-26",
"count": 1
},
{
"date": "2026-03-28",
"count": 1
},
{
"date": "2026-03-29",
"count": 3
},
{
"date": "2026-03-30",
"count": 3
},
{
"date": "2026-03-31",
"count": 1
},
{
"date": "2026-04-01",
"count": 5
},
{
"date": "2026-04-02",
"count": 2
},
{
"date": "2026-04-05",
"count": 1
},
{
"date": "2026-04-06",
"count": 2
},
{
"date": "2026-04-07",
"count": 1
},
{
"date": "2026-04-08",
"count": 2
},
{
"date": "2026-04-10",
"count": 5
},
{
"date": "2026-04-12",
"count": 2
},
{
"date": "2026-04-13",
"count": 1
},
{
"date": "2026-04-15",
"count": 2
},
{
"date": "2026-04-16",
"count": 3
},
{
"date": "2026-04-20",
"count": 1
},
{
"date": "2026-04-21",
"count": 2
},
{
"date": "2026-04-23",
"count": 2
},
{
"date": "2026-04-27",
"count": 3
},
{
"date": "2026-04-28",
"count": 2
},
{
"date": "2026-04-29",
"count": 1
},
{
"date": "2026-04-30",
"count": 2
},
{
"date": "2026-05-01",
"count": 1
},
{
"date": "2026-05-02",
"count": 1
},
{
"date": "2026-05-04",
"count": 3
},
{
"date": "2026-05-05",
"count": 1
},
{
"date": "2026-05-07",
"count": 1
},
{
"date": "2026-05-10",
"count": 1
},
{
"date": "2026-05-12",
"count": 3
},
{
"date": "2026-05-13",
"count": 5
},
{
"date": "2026-05-14",
"count": 2
},
{
"date": "2026-05-15",
"count": 1
},
{
"date": "2026-05-18",
"count": 1
},
{
"date": "2026-05-19",
"count": 1
},
{
"date": "2026-05-20",
"count": 2
},
{
"date": "2026-05-21",
"count": 2
},
{
"date": "2026-05-23",
"count": 2
},
{
"date": "2026-05-27",
"count": 1
},
{
"date": "2026-05-28",
"count": 1
},
{
"date": "2026-05-29",
"count": 1
},
{
"date": "2026-05-30",
"count": 3
},
{
"date": "2026-06-01",
"count": 2
},
{
"date": "2026-06-02",
"count": 3
},
{
"date": "2026-06-03",
"count": 2
},
{
"date": "2026-06-04",
"count": 4
},
{
"date": "2026-06-06",
"count": 1
},
{
"date": "2026-06-07",
"count": 1
},
{
"date": "2026-06-09",
"count": 3
},
{
"date": "2026-06-11",
"count": 1
},
{
"date": "2026-06-13",
"count": 1
},
{
"date": "2026-06-15",
"count": 1
},
{
"date": "2026-06-20",
"count": 1
},
{
"date": "2026-06-22",
"count": 3
},
{
"date": "2026-06-23",
"count": 1
},
{
"date": "2026-06-24",
"count": 1
},
{
"date": "2026-06-26",
"count": 1
},
{
"date": "2026-06-28",
"count": 1
},
{
"date": "2026-06-29",
"count": 3
},
{
"date": "2026-06-30",
"count": 3
},
{
"date": "2026-07-01",
"count": 1
},
{
"date": "2026-07-03",
"count": 1
},
{
"date": "2026-07-04",
"count": 1
},
{
"date": "2026-07-05",
"count": 2
},
{
"date": "2026-07-06",
"count": 3
},
{
"date": "2026-07-08",
"count": 3
},
{
"date": "2026-07-09",
"count": 3
},
{
"date": "2026-07-10",
"count": 1
},
{
"date": "2026-07-11",
"count": 2
},
{
"date": "2026-07-12",
"count": 1
},
{
"date": "2026-07-13",
"count": 2
},
{
"date": "2026-07-15",
"count": 1
},
{
"date": "2026-07-21",
"count": 2
},
{
"date": "2026-07-22",
"count": 1
},
{
"date": "2026-07-23",
"count": 1
},
{
"date": "2026-07-24",
"count": 2
}
],
"complete": true,
"collected": 572,
"total_forks": 586
},
"star_history": null,
"open_issues_and_prs": 131
},
"ai_readiness": {
"has_nix": false,
"example_dirs": [
"examples",
"recipes"
],
"has_llms_txt": false,
"has_dockerfile": false,
"has_mcp_signal": false,
"bootstrap_files": [
"Makefile",
"docs/Makefile"
],
"api_schema_files": [],
"has_devcontainer": false,
"typecheck_configs": [],
"toolchain_manifests": [],
"largest_source_bytes": 62626,
"source_files_sampled": 473,
"oversized_source_files": 1,
"agent_instruction_files": [],
"agent_instruction_max_bytes": null
},
"dependencies": {
"manifests": [
"pyproject.toml",
"setup.py"
],
"advisories": {
"error": null,
"scope": "published_package",
"source": "osv",
"findings": [
{
"name": "pillow",
"direct": true,
"version": "12.2.0",
"severity": "critical",
"ecosystem": "pypi",
"cvss_score": 9.1,
"advisory_ids": [
"GHSA-45hq-cxwh-f6vc",
"GHSA-4x4j-2g7c-83w6",
"GHSA-5x94-69rx-g8h2",
"GHSA-62p4-gmf7-7g93",
"GHSA-6r8x-57c9-28j4",
"GHSA-8v84-f9pq-wr9x",
"GHSA-9hw9-ch79-4vh6",
"GHSA-fj7v-r99m-22gq",
"GHSA-jjj6-mw9f-p565",
"GHSA-pg7v-jwj7-p798"
],
"fixed_version": "12.3.0",
"advisory_count": 26,
"oldest_advisory_days": 18
},
{
"name": "torch",
"direct": true,
"version": "2.12.0",
"severity": "moderate",
"ecosystem": "pypi",
"cvss_score": 5.3,
"advisory_ids": [
"GHSA-rrmf-rvhw-rf47"
],
"fixed_version": "2.13.0",
"advisory_count": 1,
"oldest_advisory_days": 480
}
],
"collected": true,
"malicious": [],
"truncated": false,
"by_severity": {
"critical": 1,
"moderate": 1
},
"advisory_count": 27,
"affected_count": 2,
"assessed_count": 59,
"malicious_count": 0,
"assessed_package": "pypi:llmcompressor@0.12.0",
"unassessed_count": 0,
"direct_affected_count": 2
},
"ecosystems": [
"pypi"
],
"dependencies": [],
"all_dependencies": {
"error": null,
"source": "github-sbom",
"packages": [
{
"name": "setuptools-scm",
"direct": false,
"version": "8.2.0",
"ecosystem": "pypi"
}
],
"collected": true,
"truncated": false,
"total_count": 1,
"direct_count": 0,
"indirect_count": 1
}
},
"maintainership": {
"issues": {
"open_prs": 84,
"merged_prs": 1215,
"open_issues": 47,
"closed_ratio": 0.934,
"closed_issues": 669,
"closed_unmerged_prs": 339
},
"bus_factor": 5,
"bot_contributors": 1,
"top_contributors": [
{
"type": "User",
"login": "kylesayrs",
"commits": 324,
"avatar_url": "https://avatars.githubusercontent.com/u/17103692?v=4"
},
{
"type": "User",
"login": "markurtz",
"commits": 297,
"avatar_url": "https://avatars.githubusercontent.com/u/12499095?v=4"
},
{
"type": "User",
"login": "bfineran",
"commits": 293,
"avatar_url": "https://avatars.githubusercontent.com/u/11316925?v=4"
},
{
"type": "User",
"login": "dsikka",
"commits": 276,
"avatar_url": "https://avatars.githubusercontent.com/u/22260572?v=4"
},
{
"type": "User",
"login": "rahul-tuli",
"commits": 209,
"avatar_url": "https://avatars.githubusercontent.com/u/25380596?v=4"
},
{
"type": "User",
"login": "dbogunowicz",
"commits": 117,
"avatar_url": "https://avatars.githubusercontent.com/u/97082108?v=4"
},
{
"type": "User",
"login": "KSGulin",
"commits": 111,
"avatar_url": "https://avatars.githubusercontent.com/u/66528950?v=4"
},
{
"type": "User",
"login": "natuan",
"commits": 91,
"avatar_url": "https://avatars.githubusercontent.com/u/2721633?v=4"
},
{
"type": "User",
"login": "brian-dellabetta",
"commits": 88,
"avatar_url": "https://avatars.githubusercontent.com/u/6946716?v=4"
},
{
"type": "User",
"login": "kevinaer",
"commits": 81,
"avatar_url": "https://avatars.githubusercontent.com/u/10566197?v=4"
}
],
"contributors_sampled": 99,
"top_contributor_share": 0.126
},
"quality_signals": {
"has_ci": true,
"has_tests": true,
"ci_workflows": [
"linkcheck.yml",
"linkspector.yml",
"quality-check.yaml",
"set-comment.yaml",
"stale.yml"
],
"has_docs_dir": true,
"linter_configs": [],
"has_editorconfig": false,
"has_linter_config": false,
"has_precommit_config": false
},
"security_signals": {
"lockfiles": [],
"scorecard": {
"checks": [
{
"name": "Binary-Artifacts",
"score": 10,
"reason": "no binaries found in the repo",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#binary-artifacts"
},
{
"name": "Branch-Protection",
"score": null,
"reason": "internal 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",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#branch-protection"
},
{
"name": "CI-Tests",
"score": 10,
"reason": "30 out of 30 merged PRs checked by a CI test -- score normalized to 10",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#ci-tests"
},
{
"name": "CII-Best-Practices",
"score": 0,
"reason": "no effort to earn an OpenSSF best practices badge detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#cii-best-practices"
},
{
"name": "Code-Review",
"score": 10,
"reason": "all changesets reviewed",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#code-review"
},
{
"name": "Contributors",
"score": 10,
"reason": "project has 22 contributing companies or organizations",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#contributors"
},
{
"name": "Dangerous-Workflow",
"score": 10,
"reason": "no dangerous workflow patterns detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#dangerous-workflow"
},
{
"name": "Dependency-Update-Tool",
"score": 10,
"reason": "update tool detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#dependency-update-tool"
},
{
"name": "Fuzzing",
"score": 0,
"reason": "project is not fuzzed",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#fuzzing"
},
{
"name": "License",
"score": 10,
"reason": "license file detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#license"
},
{
"name": "Maintained",
"score": 10,
"reason": "30 commit(s) and 21 issue activity found in the last 90 days -- score normalized to 10",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#maintained"
},
{
"name": "Packaging",
"score": null,
"reason": "packaging workflow not detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#packaging"
},
{
"name": "Pinned-Dependencies",
"score": 0,
"reason": "dependency not pinned by hash detected -- score normalized to 0",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#pinned-dependencies"
},
{
"name": "SAST",
"score": 0,
"reason": "SAST tool is not run on all commits -- score normalized to 0",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#sast"
},
{
"name": "Security-Policy",
"score": 0,
"reason": "security policy file not detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#security-policy"
},
{
"name": "Signed-Releases",
"score": 0,
"reason": "Project has not signed or included provenance with any releases.",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#signed-releases"
},
{
"name": "Token-Permissions",
"score": 0,
"reason": "detected GitHub workflow tokens with excessive permissions",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#token-permissions"
},
{
"name": "Vulnerabilities",
"score": 10,
"reason": "0 existing vulnerabilities detected",
"documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#vulnerabilities"
}
],
"commit": "e5eb138215726a8d00da0b308c036f77eee674f9",
"ran_at": "2026-07-25T06:48:59Z",
"aggregate_score": 5.9,
"scorecard_version": "v5.5.0"
},
"has_codeql_workflow": false,
"has_security_policy": false,
"has_dependabot_config": false
},
"contribution_flow": {
"collected": true,
"ci_last_run_at": "2026-07-24T18:22:29Z",
"oldest_open_prs": [
{
"number": 1256,
"created_at": "2025-03-15T20:55:55Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 1948,
"created_at": "2025-10-18T22:08:43Z",
"last_comment_at": "2026-07-05T03:11:43Z",
"last_comment_author": "coderabbitai"
},
{
"number": 2016,
"created_at": "2025-11-10T19:02:46Z",
"last_comment_at": "2026-07-09T18:01:30Z",
"last_comment_author": "mergify"
},
{
"number": 2027,
"created_at": "2025-11-12T12:54:56Z",
"last_comment_at": "2026-05-13T06:13:43Z",
"last_comment_author": "mergify"
},
{
"number": 2045,
"created_at": "2025-11-18T23:17:44Z",
"last_comment_at": "2025-11-18T23:17:52Z",
"last_comment_author": "github-actions"
},
{
"number": 2164,
"created_at": "2025-12-21T06:48:46Z",
"last_comment_at": "2026-05-13T02:46:19Z",
"last_comment_author": "pennyLuo-hub"
},
{
"number": 2320,
"created_at": "2026-01-31T08:53:36Z",
"last_comment_at": "2026-07-23T18:29:06Z",
"last_comment_author": "mergify"
},
{
"number": 2323,
"created_at": "2026-01-31T18:24:52Z",
"last_comment_at": "2026-07-23T18:29:14Z",
"last_comment_author": "mergify"
},
{
"number": 2378,
"created_at": "2026-02-17T20:14:14Z",
"last_comment_at": "2026-04-07T18:08:14Z",
"last_comment_author": "mergify"
},
{
"number": 2431,
"created_at": "2026-03-03T15:48:31Z",
"last_comment_at": "2026-05-30T01:08:34Z",
"last_comment_author": "mergify"
},
{
"number": 2463,
"created_at": "2026-03-10T22:33:15Z",
"last_comment_at": "2026-07-06T17:58:34Z",
"last_comment_author": "mergify"
},
{
"number": 2465,
"created_at": "2026-03-11T09:01:35Z",
"last_comment_at": "2026-05-30T01:09:38Z",
"last_comment_author": "mergify"
},
{
"number": 2492,
"created_at": "2026-03-20T02:34:23Z",
"last_comment_at": "2026-05-30T01:09:53Z",
"last_comment_author": "mergify"
},
{
"number": 2510,
"created_at": "2026-03-24T19:44:19Z",
"last_comment_at": "2026-05-13T00:14:29Z",
"last_comment_author": "mergify"
},
{
"number": 2571,
"created_at": "2026-04-06T20:01:35Z",
"last_comment_at": "2026-05-30T01:10:16Z",
"last_comment_author": "mergify"
},
{
"number": 2600,
"created_at": "2026-04-10T17:14:41Z",
"last_comment_at": "2026-05-30T01:10:28Z",
"last_comment_author": "mergify"
},
{
"number": 2602,
"created_at": "2026-04-10T17:51:58Z",
"last_comment_at": "2026-05-30T01:10:37Z",
"last_comment_author": "mergify"
},
{
"number": 2605,
"created_at": "2026-04-10T18:41:43Z",
"last_comment_at": "2026-06-15T22:38:03Z",
"last_comment_author": "mergify"
},
{
"number": 2623,
"created_at": "2026-04-16T16:55:03Z",
"last_comment_at": "2026-05-30T01:10:48Z",
"last_comment_author": "mergify"
},
{
"number": 2631,
"created_at": "2026-04-20T08:42:27Z",
"last_comment_at": "2026-04-22T21:57:01Z",
"last_comment_author": "mergify"
}
],
"last_merged_pr_at": "2026-07-24T16:29:39Z",
"ci_last_conclusion": "SUCCESS",
"oldest_open_issues": [
{
"number": 1496,
"created_at": "2025-06-01T12:19:51Z",
"last_comment_at": "2026-01-30T03:40:33Z",
"last_comment_author": "colldata79"
},
{
"number": 1850,
"created_at": "2025-09-22T20:46:06Z",
"last_comment_at": "2026-02-13T18:15:49Z",
"last_comment_author": "ArkaSanka"
},
{
"number": 1945,
"created_at": "2025-10-17T18:38:23Z",
"last_comment_at": null,
"last_comment_author": null
},
{
"number": 1966,
"created_at": "2025-10-27T19:23:57Z",
"last_comment_at": "2026-05-07T14:21:34Z",
"last_comment_author": "dsikka"
},
{
"number": 2006,
"created_at": "2025-11-09T19:40:10Z",
"last_comment_at": "2025-11-10T11:12:59Z",
"last_comment_author": "mratsim"
},
{
"number": 2031,
"created_at": "2025-11-13T22:26:46Z",
"last_comment_at": "2026-03-12T16:44:41Z",
"last_comment_author": "HDCharles"
},
{
"number": 2296,
"created_at": "2026-01-27T17:05:37Z",
"last_comment_at": "2026-07-21T18:18:41Z",
"last_comment_author": "github-actions"
},
{
"number": 2479,
"created_at": "2026-03-17T14:01:05Z",
"last_comment_at": "2026-06-30T18:59:37Z",
"last_comment_author": "github-actions"
},
{
"number": 2480,
"created_at": "2026-03-17T15:53:19Z",
"last_comment_at": "2026-07-07T18:58:08Z",
"last_comment_author": "ishrith-gowda"
},
{
"number": 2490,
"created_at": "2026-03-19T18:53:56Z",
"last_comment_at": "2026-07-20T18:57:35Z",
"last_comment_author": "github-actions"
},
{
"number": 2494,
"created_at": "2026-03-20T07:26:40Z",
"last_comment_at": "2026-07-02T20:56:04Z",
"last_comment_author": "kylesayrs"
},
{
"number": 2509,
"created_at": "2026-03-24T17:36:11Z",
"last_comment_at": "2026-06-24T18:54:26Z",
"last_comment_author": "github-actions"
},
{
"number": 2512,
"created_at": "2026-03-25T01:49:48Z",
"last_comment_at": "2026-07-02T21:10:02Z",
"last_comment_author": "kylesayrs"
},
{
"number": 2522,
"created_at": "2026-03-26T13:42:58Z",
"last_comment_at": "2026-05-04T16:29:19Z",
"last_comment_author": "brian-dellabetta"
},
{
"number": 2646,
"created_at": "2026-04-24T14:23:55Z",
"last_comment_at": "2026-05-05T06:21:37Z",
"last_comment_author": "jwpark33"
},
{
"number": 2654,
"created_at": "2026-04-26T00:21:59Z",
"last_comment_at": "2026-07-07T04:30:07Z",
"last_comment_author": "kylesayrs"
},
{
"number": 2656,
"created_at": "2026-04-26T14:05:26Z",
"last_comment_at": "2026-04-28T19:52:34Z",
"last_comment_author": "brian-dellabetta"
},
{
"number": 2667,
"created_at": "2026-04-29T19:19:09Z",
"last_comment_at": "2026-04-29T21:50:06Z",
"last_comment_author": "jayakumarpujar"
},
{
"number": 2690,
"created_at": "2026-05-07T14:43:40Z",
"last_comment_at": "2026-05-15T17:59:08Z",
"last_comment_author": "HDCharles"
},
{
"number": 2693,
"created_at": "2026-05-10T05:05:07Z",
"last_comment_at": "2026-06-15T07:35:37Z",
"last_comment_author": "Tobi-Adesoye"
}
]
}
},
"config": {
"disabled_metrics": [],
"disabled_categories": [],
"disabled_components": {}
},
"source": {
"url": "https://github.com/vllm-project/llm-compressor",
"host": "github.com",
"name": "llm-compressor",
"owner": "vllm-project"
},
"metrics": {
"overall": {
"key": "overall",
"band": "good",
"name": "Overall health",
"note": null,
"notes": [],
"value": 84,
"inputs": {
"security": 61,
"vitality": 95,
"community": 90,
"governance": 87,
"engineering": 81
},
"components": []
},
"categories": [
{
"key": "vitality",
"band": "excellent",
"name": "Vitality",
"value": 95,
"weight": 0.22,
"metrics": [
{
"key": "development_activity",
"band": "excellent",
"name": "Development activity",
"note": null,
"notes": [],
"value": 99,
"inputs": {
"commits_last_year": 654,
"human_commit_share": 1,
"days_since_last_push": 0,
"active_weeks_last_year": 50
},
"components": [
{
"key": "push_recency",
"name": "Push recency",
"detail": "last push 0 days ago",
"points": 36,
"status": "met",
"details": [
{
"code": "push_recency",
"params": {
"days": 0
}
}
],
"max_points": 36
},
{
"key": "commit_cadence",
"name": "Commit cadence",
"detail": "50/52 weeks with commits",
"points": 34.6,
"status": "partial",
"details": [
{
"code": "commit_cadence_weeks",
"params": {
"weeks": 50
}
}
],
"max_points": 36
},
{
"key": "commit_volume",
"name": "Commit volume",
"detail": "654 commits in the last year",
"points": 18,
"status": "met",
"details": [
{
"code": "commits_last_year",
"params": {
"count": 654
}
}
],
"max_points": 18
},
{
"key": "openssf_scorecard_maintained",
"name": "OpenSSF Scorecard: Maintained",
"detail": "30 commit(s) and 21 issue activity found in the last 90 days -- score normalized to 10",
"points": 10,
"status": "met",
"details": [],
"max_points": 10
}
]
},
{
"key": "release_discipline",
"band": "excellent",
"name": "Release discipline",
"note": null,
"notes": [],
"value": 90,
"inputs": {
"releases_count": 27,
"latest_release_tag": "0.7.1.3",
"releases_from_tags": false,
"days_since_latest_release": 29,
"mean_days_between_releases": 14.7
},
"components": [
{
"key": "ships_releases",
"name": "Ships releases",
"detail": "27 releases published",
"points": 27,
"status": "met",
"details": [
{
"code": "releases_published",
"params": {
"count": 27
}
}
],
"max_points": 27
},
{
"key": "release_recency",
"name": "Release recency",
"detail": "latest release 29 days ago",
"points": 36,
"status": "met",
"details": [
{
"code": "release_recency",
"params": {
"days": 29
}
}
],
"max_points": 36
},
{
"key": "release_cadence",
"name": "Release cadence",
"detail": "a release every ~14.7 days",
"points": 27,
"status": "met",
"details": [
{
"code": "release_cadence",
"params": {
"gap": 14.7
}
}
],
"max_points": 27
},
{
"key": "openssf_scorecard_signed_releases",
"name": "OpenSSF Scorecard: Signed-Releases",
"detail": "Project has not signed or included provenance with any releases.",
"points": 0,
"status": "missed",
"details": [],
"max_points": 10
}
]
},
{
"key": "abandonment",
"band": "excellent",
"name": "Abandonment",
"note": null,
"notes": [],
"value": 100,
"inputs": {
"cap": null,
"state": "maintained",
"guards": [],
"signals": [],
"red_flag": false,
"multiplier_pct": 100,
"declared_reason": null,
"unverified_reason": null,
"unanswered_open_prs": null,
"unanswered_open_issues": null,
"days_since_last_merged_pr": null,
"days_since_last_human_commit": 0,
"days_since_last_human_commit_is_floor": false
},
"components": [
{
"key": "project_is_still_maintained",
"name": "Project is still maintained",
"detail": "last human commit 0 days ago",
"points": 100,
"status": "met",
"details": [
{
"code": "abandonment_maintained",
"params": {
"days": 0
}
}
],
"max_points": 100
}
]
}
],
"description": "Is the project alive — is code being written and are releases shipping?"
},
{
"key": "community",
"band": "excellent",
"name": "Community & Adoption",
"value": 90,
"weight": 0.18,
"metrics": [
{
"key": "popularity",
"band": "excellent",
"name": "Popularity & adoption",
"note": null,
"notes": [],
"value": 89,
"inputs": {
"forks": 586,
"stars": 3580,
"watchers": 29,
"growth_state": "unverified",
"growth_factor_pct": 100,
"growth_unverified_reason": "no_history"
},
"components": [
{
"key": "stars",
"name": "Stars",
"detail": "3,580 stars",
"points": 57.6,
"status": "partial",
"details": [
{
"code": "stars",
"params": {
"count": 3580
}
}
],
"max_points": 60
},
{
"key": "forks",
"name": "Forks",
"detail": "586 forks",
"points": 23.1,
"status": "partial",
"details": [
{
"code": "forks",
"params": {
"count": 586
}
}
],
"max_points": 25
},
{
"key": "watchers",
"name": "Watchers",
"detail": "29 watchers",
"points": 8,
"status": "partial",
"details": [
{
"code": "watchers",
"params": {
"count": 29
}
}
],
"max_points": 15
}
]
},
{
"key": "community_health",
"band": "excellent",
"name": "Community health",
"note": null,
"notes": [],
"value": 92,
"inputs": {
"has_readme": true,
"has_license": true,
"has_contributing": true,
"has_issue_template": false,
"has_code_of_conduct": true,
"has_pull_request_template": true
},
"components": [
{
"key": "readme",
"name": "README",
"detail": null,
"points": 22.5,
"status": "met",
"details": [],
"max_points": 22.5
},
{
"key": "license",
"name": "License",
"detail": "recognized license (Apache-2.0)",
"points": 22.5,
"status": "met",
"details": [
{
"code": "license_standard",
"params": {}
},
{
"code": "license_spdx",
"params": {
"spdx": "Apache-2.0"
}
}
],
"max_points": 22.5
},
{
"key": "contributing_guide",
"name": "CONTRIBUTING guide",
"detail": null,
"points": 18,
"status": "met",
"details": [],
"max_points": 18
},
{
"key": "code_of_conduct",
"name": "Code of conduct",
"detail": null,
"points": 13.5,
"status": "met",
"details": [],
"max_points": 13.5
},
{
"key": "issue_template",
"name": "Issue template",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 7.2
},
{
"key": "pr_template",
"name": "PR template",
"detail": null,
"points": 6.3,
"status": "met",
"details": [],
"max_points": 6.3
}
]
},
{
"key": "ecosystem_adoption",
"band": "excellent",
"name": "Ecosystem adoption (downloads)",
"note": "Excluded from scoring (no data or not applicable): Registry dependents. Remaining weights renormalized.",
"notes": [
{
"code": "excluded_no_data",
"params": {
"components": [
"registry_dependents"
]
}
},
{
"code": "weights_renormalized",
"params": {}
}
],
"value": 88,
"inputs": {
"packages": [
"llmcompressor"
],
"dependents": null,
"ecosystems": "pypi",
"total_downloads": null,
"monthly_downloads": 190161
},
"components": [
{
"key": "monthly_downloads",
"name": "Monthly downloads",
"detail": "190,161 downloads/month across pypi",
"points": 70.4,
"status": "partial",
"details": [
{
"code": "downloads_monthly",
"params": {
"count": 190161,
"ecosystems": "pypi"
}
}
],
"max_points": 80
},
{
"key": "registry_dependents",
"name": "Registry dependents",
"detail": "not reported by this ecosystem",
"points": 0,
"status": "excluded",
"details": [
{
"code": "not_reported_by_this_ecosystem",
"params": {}
}
],
"max_points": 20
}
]
}
],
"description": "Does the project have users, downloads, attention, and a welcoming setup for contributors?"
},
{
"key": "governance",
"band": "excellent",
"name": "Sustainability & Governance",
"value": 87,
"weight": 0.24,
"metrics": [
{
"key": "maintainer_resilience",
"band": "excellent",
"name": "Maintainer resilience (bus factor)",
"note": null,
"notes": [],
"value": 89,
"inputs": {
"bus_factor": 5,
"contributors_sampled": 99,
"top_contributor_share": 0.126
},
"components": [
{
"key": "bus_factor",
"name": "Bus factor",
"detail": "5 contributor(s) cover half of all commits",
"points": 45.9,
"status": "partial",
"details": [
{
"code": "bus_factor",
"params": {
"count": 5
}
}
],
"max_points": 54
},
{
"key": "commit_distribution",
"name": "Commit distribution",
"detail": "top contributor authored 13% of commits",
"points": 19.7,
"status": "partial",
"details": [
{
"code": "top_contributor_share",
"params": {
"share": 13
}
}
],
"max_points": 22.5
},
{
"key": "contributor_breadth",
"name": "Contributor breadth",
"detail": "99 contributors",
"points": 13.5,
"status": "met",
"details": [
{
"code": "contributors_sampled",
"params": {
"count": 99
}
}
],
"max_points": 13.5
},
{
"key": "openssf_scorecard_contributors",
"name": "OpenSSF Scorecard: Contributors",
"detail": "project has 22 contributing companies or organizations",
"points": 10,
"status": "met",
"details": [],
"max_points": 10
}
]
},
{
"key": "responsiveness",
"band": "excellent",
"name": "Issue & PR responsiveness",
"note": null,
"notes": [],
"value": 89,
"inputs": {
"merged_prs": 1215,
"open_issues": 47,
"closed_issues": 669,
"issue_closed_ratio": 0.934,
"closed_unmerged_prs": 339
},
"components": [
{
"key": "issue_resolution",
"name": "Issue resolution",
"detail": "93% of issues closed",
"points": 43.7,
"status": "partial",
"details": [
{
"code": "issues_closed_share",
"params": {
"share": 93
}
}
],
"max_points": 46.75
},
{
"key": "pr_acceptance",
"name": "PR acceptance",
"detail": "1215/1554 decided PRs merged",
"points": 29.9,
"status": "partial",
"details": [
{
"code": "decided_prs_merged",
"params": {
"merged": 1215,
"decided": 1554
}
}
],
"max_points": 38.25
},
{
"key": "openssf_scorecard_code_review",
"name": "OpenSSF Scorecard: Code-Review",
"detail": "all changesets reviewed",
"points": 15,
"status": "met",
"details": [],
"max_points": 15
}
]
},
{
"key": "stewardship",
"band": "good",
"name": "Ownership & stewardship",
"note": null,
"notes": [],
"value": 73,
"inputs": {
"followers": 3752,
"owner_type": "Organization",
"is_verified": null,
"owner_login": "vllm-project",
"public_repos": 45,
"account_age_days": 1132
},
"components": [
{
"key": "ownership_backing",
"name": "Ownership backing",
"detail": "organization-owned",
"points": 30,
"status": "met",
"details": [
{
"code": "owner_organization",
"params": {}
}
],
"max_points": 30
},
{
"key": "verified_domain",
"name": "Verified domain",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 20
},
{
"key": "owner_reach",
"name": "Owner reach",
"detail": "3,752 followers of vllm-project",
"points": 25,
"status": "met",
"details": [
{
"code": "owner_followers",
"params": {
"count": 3752,
"login": "vllm-project"
}
}
],
"max_points": 25
},
{
"key": "track_record",
"name": "Track record",
"detail": "45 public repos, account ~3 yr old",
"points": 18.3,
"status": "partial",
"details": [
{
"code": "public_repos",
"params": {
"count": 45
}
},
{
"code": "account_age_years",
"params": {
"years": 3
}
}
],
"max_points": 25
}
]
},
{
"key": "package_maintenance",
"band": "excellent",
"name": "Package maintenance",
"note": null,
"notes": [],
"value": 100,
"inputs": {
"packages": [
"llmcompressor"
],
"ecosystems": "pypi",
"any_deprecated": false,
"min_days_since_publish": 39
},
"components": [
{
"key": "published_resolvable",
"name": "Published & resolvable",
"detail": "1 package(s) on pypi",
"points": 25,
"status": "met",
"details": [
{
"code": "packages_published",
"params": {
"count": 1,
"ecosystems": "pypi"
}
}
],
"max_points": 25
},
{
"key": "publish_recency",
"name": "Publish recency",
"detail": "latest publish 39 days ago",
"points": 35,
"status": "met",
"details": [
{
"code": "publish_recency",
"params": {
"days": 39
}
}
],
"max_points": 35
},
{
"key": "version_history",
"name": "Version history",
"detail": "77 published versions",
"points": 20,
"status": "met",
"details": [
{
"code": "published_versions",
"params": {
"count": 77
}
}
],
"max_points": 20
},
{
"key": "not_deprecated",
"name": "Not deprecated",
"detail": "active, not deprecated or yanked",
"points": 20,
"status": "met",
"details": [
{
"code": "package_not_deprecated",
"params": {}
}
],
"max_points": 20
}
]
}
],
"description": "Will the project survive its people — bus factor, responsiveness, who backs it, and package upkeep?"
},
{
"key": "engineering",
"band": "good",
"name": "Engineering Quality",
"value": 81,
"weight": 0.2,
"metrics": [
{
"key": "engineering_practices",
"band": "moderate",
"name": "Engineering practices",
"note": null,
"notes": [],
"value": 68,
"inputs": {
"has_ci": true,
"has_tests": true,
"has_editorconfig": false,
"has_linter_config": false,
"has_precommit_config": false
},
"components": [
{
"key": "ci_workflows",
"name": "CI workflows",
"detail": "5 workflow(s)",
"points": 24,
"status": "met",
"details": [
{
"code": "ci_workflows",
"params": {
"count": 5
}
}
],
"max_points": 24
},
{
"key": "tests_present",
"name": "Tests present",
"detail": null,
"points": 24,
"status": "met",
"details": [],
"max_points": 24
},
{
"key": "linter_config",
"name": "Linter config",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 16
},
{
"key": "pre_commit_hooks",
"name": "Pre-commit hooks",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 9.6
},
{
"key": "editorconfig",
"name": ".editorconfig",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 6.4
},
{
"key": "openssf_scorecard_ci_tests",
"name": "OpenSSF Scorecard: CI-Tests",
"detail": "30 out of 30 merged PRs checked by a CI test -- score normalized to 10",
"points": 20,
"status": "met",
"details": [],
"max_points": 20
}
]
},
{
"key": "documentation",
"band": "excellent",
"name": "Documentation",
"note": null,
"notes": [],
"value": 100,
"inputs": {
"topics": [
"compression",
"quantization"
],
"has_wiki": true,
"homepage": "https://docs.vllm.ai/projects/llm-compressor",
"has_readme": true,
"has_docs_dir": true,
"has_description": true
},
"components": [
{
"key": "readme",
"name": "README",
"detail": null,
"points": 30,
"status": "met",
"details": [],
"max_points": 30
},
{
"key": "documentation_directory",
"name": "Documentation directory",
"detail": null,
"points": 25,
"status": "met",
"details": [],
"max_points": 25
},
{
"key": "documentation_homepage_site",
"name": "Documentation / homepage site",
"detail": "https://docs.vllm.ai/projects/llm-compressor",
"points": 15,
"status": "met",
"details": [],
"max_points": 15
},
{
"key": "repository_description",
"name": "Repository description",
"detail": null,
"points": 10,
"status": "met",
"details": [],
"max_points": 10
},
{
"key": "topics",
"name": "Topics",
"detail": "2 topics",
"points": 10,
"status": "met",
"details": [
{
"code": "topics_count",
"params": {
"count": 2
}
}
],
"max_points": 10
},
{
"key": "wiki",
"name": "Wiki",
"detail": null,
"points": 10,
"status": "met",
"details": [],
"max_points": 10
}
]
}
],
"description": "Are baseline engineering and documentation practices in place?"
},
{
"key": "security",
"band": "moderate",
"name": "Security",
"value": 61,
"weight": 0.16,
"metrics": [
{
"key": "security_posture",
"band": "moderate",
"name": "Security posture",
"note": "Excluded from scoring (no data or not applicable): Branch-Protection, Packaging. Remaining weights renormalized.",
"notes": [
{
"code": "excluded_no_data",
"params": {
"components": [
"branch_protection",
"packaging"
]
}
},
{
"code": "weights_renormalized",
"params": {}
}
],
"value": 59,
"inputs": {
"source": "openssf_scorecard",
"checks_evaluated": 16,
"scorecard_version": "v5.5.0",
"checks_inconclusive": 2,
"scorecard_aggregate": 5.9
},
"components": [
{
"key": "binary_artifacts",
"name": "Binary-Artifacts",
"detail": "no binaries found in the repo",
"points": 7.5,
"status": "met",
"details": [],
"max_points": 7.5
},
{
"key": "branch_protection",
"name": "Branch-Protection",
"detail": "internal 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",
"points": 0,
"status": "excluded",
"details": [
{
"code": "no_data",
"params": {}
}
],
"max_points": 7.5
},
{
"key": "ci_tests",
"name": "CI-Tests",
"detail": "30 out of 30 merged PRs checked by a CI test -- score normalized to 10",
"points": 2.5,
"status": "met",
"details": [],
"max_points": 2.5
},
{
"key": "cii_best_practices",
"name": "CII-Best-Practices",
"detail": "no effort to earn an OpenSSF best practices badge detected",
"points": 0,
"status": "missed",
"details": [],
"max_points": 2.5
},
{
"key": "code_review",
"name": "Code-Review",
"detail": "all changesets reviewed",
"points": 7.5,
"status": "met",
"details": [],
"max_points": 7.5
},
{
"key": "contributors",
"name": "Contributors",
"detail": "project has 22 contributing companies or organizations",
"points": 2.5,
"status": "met",
"details": [],
"max_points": 2.5
},
{
"key": "dangerous_workflow",
"name": "Dangerous-Workflow",
"detail": "no dangerous workflow patterns detected",
"points": 10,
"status": "met",
"details": [],
"max_points": 10
},
{
"key": "dependency_update_tool",
"name": "Dependency-Update-Tool",
"detail": "update tool detected",
"points": 7.5,
"status": "met",
"details": [],
"max_points": 7.5
},
{
"key": "fuzzing",
"name": "Fuzzing",
"detail": "project is not fuzzed",
"points": 0,
"status": "missed",
"details": [],
"max_points": 5
},
{
"key": "license",
"name": "License",
"detail": "license file detected",
"points": 2.5,
"status": "met",
"details": [],
"max_points": 2.5
},
{
"key": "maintained",
"name": "Maintained",
"detail": "30 commit(s) and 21 issue activity found in the last 90 days -- score normalized to 10",
"points": 7.5,
"status": "met",
"details": [],
"max_points": 7.5
},
{
"key": "packaging",
"name": "Packaging",
"detail": "packaging workflow not detected",
"points": 0,
"status": "excluded",
"details": [
{
"code": "no_data",
"params": {}
}
],
"max_points": 5
},
{
"key": "pinned_dependencies",
"name": "Pinned-Dependencies",
"detail": "dependency not pinned by hash detected -- score normalized to 0",
"points": 0,
"status": "missed",
"details": [],
"max_points": 5
},
{
"key": "sast",
"name": "SAST",
"detail": "SAST tool is not run on all commits -- score normalized to 0",
"points": 0,
"status": "missed",
"details": [],
"max_points": 5
},
{
"key": "security_policy",
"name": "Security-Policy",
"detail": "security policy file not detected",
"points": 0,
"status": "missed",
"details": [],
"max_points": 5
},
{
"key": "signed_releases",
"name": "Signed-Releases",
"detail": "Project has not signed or included provenance with any releases.",
"points": 0,
"status": "missed",
"details": [],
"max_points": 7.5
},
{
"key": "token_permissions",
"name": "Token-Permissions",
"detail": "detected GitHub workflow tokens with excessive permissions",
"points": 0,
"status": "missed",
"details": [],
"max_points": 7.5
},
{
"key": "vulnerabilities",
"name": "Vulnerabilities",
"detail": "0 existing vulnerabilities detected",
"points": 7.5,
"status": "met",
"details": [],
"max_points": 7.5
}
]
},
{
"key": "dependency_advisories",
"band": "moderate",
"name": "Dependency advisories",
"note": "Matched the pypi:llmcompressor@0.12.0 runtime dependency closure — what installing the published package pulls in — 59 packages. Reachability is not analyzed.",
"notes": [
{
"code": "advisories_scope_published",
"params": {
"package": "pypi:llmcompressor@0.12.0",
"assessed": 59
}
},
{
"code": "advisories_reachability",
"params": {}
}
],
"value": 69,
"inputs": {
"source": "osv",
"advisories": 27,
"affected_packages": 2,
"assessed_packages": 59,
"unassessed_packages": 0,
"affected_by_severity": "critical 1, moderate 1",
"direct_affected_packages": 2
},
"components": [
{
"key": "direct_dependencies_free_of_known_advisories",
"name": "Direct dependencies free of known advisories",
"detail": "2 affected: pillow 12.2.0 (critical 9.1), torch 2.12.0 (moderate 5.3)",
"points": 8.4,
"status": "partial",
"details": [
{
"code": "advisories_affected",
"params": {
"count": 2,
"packages": "pillow 12.2.0 (critical 9.1), torch 2.12.0 (moderate 5.3)"
}
}
],
"max_points": 35
},
{
"key": "indirect_dependencies_free_of_known_advisories",
"name": "Indirect dependencies free of known advisories",
"detail": "no indirect dependency carries a known advisory",
"points": 25,
"status": "met",
"details": [
{
"code": "no_indirect_advisories",
"params": {}
}
],
"max_points": 25
},
{
"key": "no_advisories_left_outstanding",
"name": "No advisories left outstanding",
"detail": "1 advisory-carrying package(s) unaddressed past 90 days; oldest published 480 days ago",
"points": 35.3,
"status": "partial",
"details": [
{
"code": "advisories_stale",
"params": {
"days": 90,
"count": 1,
"oldest": 480
}
}
],
"max_points": 40
}
]
},
{
"key": "malicious_dependencies",
"band": "excellent",
"name": "Malicious dependencies",
"note": null,
"notes": [],
"value": 100,
"inputs": {
"source": "osv",
"meaning": "reported as a malicious package by the OpenSSF corpus; the remedy is removal or moving off the compromised name, never an upgrade of the same artifact. Versions the registry has since pulled are listed but not scored",
"packages": [],
"red_flag": false,
"assessed_packages": 59,
"malicious_packages": 0,
"direct_malicious_packages": 0,
"withdrawn_malicious_packages": 0,
"installable_malicious_packages": 0
},
"components": [
{
"key": "no_dependency_reported_as_a_malicious_package",
"name": "No dependency reported as a malicious package",
"detail": "no dependency is reported as a malicious package",
"points": 100,
"status": "met",
"details": [
{
"code": "no_malicious_dependencies",
"params": {}
}
],
"max_points": 100
}
]
},
{
"key": "high_risk_jurisdiction_exposure",
"band": "excellent",
"name": "High-Risk Jurisdiction Exposure",
"note": "Only high-confidence self-published location evidence affects this multiplier. Ambiguous matches are review-only; country evidence is not proof of nationality, citizenship, legal registration, malicious intent, or sanctions status.",
"notes": [
{
"code": "jurisdiction_evidence_limits",
"params": {}
}
],
"value": 100,
"inputs": {
"meaning": "self-published location evidence; not nationality or citizenship",
"red_flag": false,
"exposures": [],
"policy_countries": [
"Russia",
"Iran",
"North Korea"
],
"review_only_matches": 0,
"assessed_self_published_locations": 9
},
"components": [
{
"key": "policy_exposure_multiplier",
"name": "Policy exposure multiplier",
"detail": "no confirmed policy-scope location match",
"points": 100,
"status": "met",
"details": [
{
"code": "jurisdiction_no_match",
"params": {}
}
],
"max_points": 100
}
]
}
],
"description": "Are visible security and supply-chain practices strong, with no malicious dependency and no unresolved high-risk jurisdiction exposure?"
},
{
"key": "ai_readiness",
"band": "at_risk",
"name": "AI Readiness",
"value": 46,
"weight": 0,
"metrics": [
{
"key": "ai_agent_context",
"band": "at_risk",
"name": "Agent context & guidance",
"note": null,
"notes": [],
"value": 40,
"inputs": {
"has_llms_txt": false,
"legible_history_share": 1,
"agent_instruction_files": [],
"agent_instruction_max_bytes": null
},
"components": [
{
"key": "agent_instructions",
"name": "Agent instructions",
"detail": "no CLAUDE.md / AGENTS.md / editor rules",
"points": 0,
"status": "missed",
"details": [
{
"code": "no_agent_instructions",
"params": {}
}
],
"max_points": 45
},
{
"key": "machine_readable_docs_llms_txt",
"name": "Machine-readable docs (llms.txt)",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 15
},
{
"key": "legible_commit_history",
"name": "Legible commit history",
"detail": "100 of 100 human commits state their intent (structured subject or explanatory body)",
"points": 40,
"status": "met",
"details": [
{
"code": "legible_history",
"params": {
"legible": 100,
"sampled": 100
}
}
],
"max_points": 40
}
]
},
{
"key": "ai_verify_loop",
"band": "moderate",
"name": "Verify loop (build / test / typecheck)",
"note": null,
"notes": [],
"value": 50,
"inputs": {
"has_nix": false,
"has_tests": true,
"lockfiles": [],
"has_dockerfile": false,
"typed_language": false,
"bootstrap_files": [
"Makefile",
"docs/Makefile"
],
"has_devcontainer": false,
"has_linter_config": false,
"typecheck_configs": [],
"agent_commit_share": 0.26,
"toolchain_manifests": [],
"dependency_bot_commit_share": 0
},
"components": [
{
"key": "one_command_bootstrap",
"name": "One-command bootstrap",
"detail": "Makefile, docs/Makefile",
"points": 18,
"status": "met",
"details": [
{
"code": "file_list",
"params": {
"files": "Makefile, docs/Makefile"
}
}
],
"max_points": 18
},
{
"key": "automated_tests",
"name": "Automated tests",
"detail": null,
"points": 22,
"status": "met",
"details": [],
"max_points": 22
},
{
"key": "lint_format_config",
"name": "Lint / format config",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 11
},
{
"key": "static_type_checking",
"name": "Static type checking",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 11
},
{
"key": "reproducible_environment",
"name": "Reproducible environment",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 10
},
{
"key": "demonstrated_agent_practice",
"name": "Demonstrated agent practice",
"detail": "26 of the last 100 commits agent-authored or agent-credited",
"points": 10,
"status": "met",
"details": [
{
"code": "agent_authored_commits",
"params": {
"count": 26,
"sampled": 100
}
}
],
"max_points": 10
},
{
"key": "automated_maintenance",
"name": "Automated maintenance",
"detail": "no automated dependency updates observed",
"points": 0,
"status": "missed",
"details": [
{
"code": "no_dependency_automation",
"params": {}
}
],
"max_points": 8
},
{
"key": "openssf_scorecard_pinned_dependencies",
"name": "OpenSSF Scorecard: Pinned-Dependencies",
"detail": "dependency not pinned by hash detected -- score normalized to 0",
"points": 0,
"status": "missed",
"details": [],
"max_points": 10
}
]
},
{
"key": "ai_code_legibility",
"band": "moderate",
"name": "Code legibility for models",
"note": null,
"notes": [],
"value": 55,
"inputs": {
"primary_language": "Python",
"largest_source_bytes": 62626,
"source_files_sampled": 473,
"oversized_source_files": 1
},
"components": [
{
"key": "type_checkable_code",
"name": "Type-checkable code",
"detail": "Python without a type-check config",
"points": 0,
"status": "missed",
"details": [
{
"code": "no_typecheck_config_language",
"params": {
"language": "Python"
}
}
],
"max_points": 45
},
{
"key": "manageable_file_sizes",
"name": "Manageable file sizes",
"detail": "1/473 source files over 60KB",
"points": 54.9,
"status": "partial",
"details": [
{
"code": "oversized_source_files",
"params": {
"kb": 60,
"sampled": 473,
"oversized": 1
}
}
],
"max_points": 55
}
]
},
{
"key": "ai_interfaces",
"band": "at_risk",
"name": "Machine-readable interfaces",
"note": null,
"notes": [],
"value": 40,
"inputs": {
"example_dirs": [
"examples",
"recipes"
],
"has_mcp_signal": false,
"api_schema_files": []
},
"components": [
{
"key": "api_schema_openapi_graphql_proto",
"name": "API schema (OpenAPI/GraphQL/proto)",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 40
},
{
"key": "mcp_server",
"name": "MCP server",
"detail": null,
"points": 0,
"status": "missed",
"details": [],
"max_points": 20
},
{
"key": "runnable_examples",
"name": "Runnable examples",
"detail": "examples, recipes",
"points": 40,
"status": "met",
"details": [
{
"code": "file_list",
"params": {
"files": "examples, recipes"
}
}
],
"max_points": 40
}
]
}
],
"description": "How well is the repo equipped to be developed and maintained with AI coding agents? An independent, experimental badge — weight 0.0, so it is surfaced on its own and does not affect the overall health score."
}
],
"metrics_version": "1.13.0"
},
"warnings": [
"Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token"
],
"report_type": "repository",
"generated_at": "2026-07-25T06:49:29.011377Z",
"schema_version": "0.27.0",
"badge_url": "https://raw.githubusercontent.com/inspect-software/badges/main/v1/v/vllm-project/llm-compressor.svg",
"full_name": "vllm-project/llm-compressor",
"license_state": "standard",
"license_spdx": "Apache-2.0"
}
Verwandte Einträge
Geprüfte Repositories, die Katalog-Tags oder Ökosysteme mit vllm-project/llm-compressor teilen. Nebeneinander lesen →
Bewertungen sind Signale, keine Garantien. Sie spiegeln öffentlich sichtbare Praxis auf GitHub wider — kein Code-Audit und keine Sicherheitsgarantie.
Fehlende Daten werden ausgeschlossen und die Gewichte neu normiert, nie als null bewertet. Die Methodik ist versioniert und offen: Metriken v1.13.0, Schema v0.27.0 — vollständige Methodik · Metriken-Wiki.
