CVE-2026-100653

MEDIUMPre-NVD 6.56.5—
EchelonGraph scoreHIGH confidence

Score 6.5 from GitHub Security Advisory (severity: HIGH) published 2026-09-26. The CNA's CVSS baseline 6.5; sources differ by 0.0.

Triggered by: GitHub Security Advisory CVSS
Sources: cna:vulncheck, epss, ghsa
Trending — 3 sources updated this weekElevated
6.5EG
EchelonGraph verdictMonitorLow exploitation likelihood right now — keep watching.
  • Lower severity and no public exploit yet
CISA-KEV: Not listedEPSS PROB: 0.3%CVSS: 6.5Exploit: Elevated riskExposed services: Not assessed

A fix is available — apply it.

vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.

CVSS v3
6.5
EG Score
6.5MEDIUMhigh confidence
EG Risk
45
EG Risk 45/100CISA SSVC

EG Risk is EchelonGraph's 0–100 priority score: it fuses intrinsic severity with real-world exploitation and automatability so you can rank equal-severity CVEs and fix the most dangerous first. Higher = act sooner. Distinct from the 0–10 EG Score (severity).

How it’s computed
Severity65% × 45%
Exploitation40% × 40%
Automatability0% × 15%
CISA SSVC: Track at low or medium mission impact; Track* at high (mission-essential systems).
Action: A fix is available. Apply it within your standard update timelines.
EPSS PROB
0.3%
EPSS %ILE
19th
KEV
Not listed

CISA SSVCTrack at low or medium mission impact; Track* at high (mission-essential systems).

A fix is available. Apply it within your standard update timelines.

Exploitation public PoC (CISA Vulnrichment) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). Mission impact is CISA's Mission & Well-being decision point, and only you can judge it: high means the affected system is essential to your organisation's mission, or its compromise could cause irreversible harm to people. CISA's decision table

Published

September 26, 2026

Last Modified

October 1, 2026

Advisory Details (3)

Auto-updated Sep 26, 2026
🔬 Proof of concept available. Patch available. Sources: github, github_commit.
generic Patch Available

vLLM 0.22.1 before 0.28.0 Incomplete Artifact Pin Propagation | Advisories | VulnCheck

https://www.vulncheck.com/advisories/vllm-0.22.1-before-0.28.0-incomplete-artifact-pin-propagation
github_commit

commit d26a28ab0336 (vllm-project/vllm)

Fix landed in vllm-project/vllm commit d26a28ab0336 — awaiting tagged release

https://github.com/vllm-project/vllm/commit/d26a28ab033697f55a1414b5b0435de7cd6045b6
github Patch Available🟡 PoC Available

Incomplete artifact pin propagation in FunAudioChat and Tarsier2 · Advisory · vllm-project/vllm · GitHub

https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q

Vendor Advisories for CVE-2026-100653(1)

These vendors published their own advisory mentioning this CVE — often with vendor-specific remediation steps + affected product lists not in NVD.

Weakness Classification(1)

MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.

Data Freshness Timeline

(refreshed 22× in last 7d / 27× in last 30d)

Each row is a source pipeline that fetched or updated this CVE on that date, with what changed. For example, "NVD update" means NVD published or revised its analysis for this CVE; "MITRE cvelistV5" means we ingested or refreshed it from the CNA feed. Most recent first.

  1. 2026-10-04 10:37 UTCEG score recompute
  2. 2026-10-04 10:37 UTCGHSA enrichment
  3. 2026-10-03 14:25 UTCEPSS rescore
  4. 2026-10-03 10:50 UTCEG score recompute
  5. 2026-10-03 10:50 UTCGHSA enrichment
  6. 2026-10-02 11:05 UTCEG score recompute
  7. 2026-10-02 11:05 UTCGHSA enrichment
  8. 2026-10-01 19:49 UTCEPSS rescore
  9. 2026-10-01 11:21 UTCEG score recompute
  10. 2026-10-01 11:21 UTCGHSA enrichment
  11. 2026-09-30 18:41 UTCEG score recompute
  12. 2026-09-30 18:41 UTCGHSA enrichment
  13. 2026-09-30 17:55 UTCEG score recompute
  14. 2026-09-30 17:55 UTCGHSA enrichment
  15. 2026-09-30 15:02 UTCEPSS rescore
  16. 2026-09-29 23:48 UTCEG score recompute
  17. 2026-09-29 23:48 UTCGHSA enrichment
  18. 2026-09-29 15:20 UTCEPSS rescore
  19. 2026-09-28 20:50 UTCGHSA enrichment
  20. 2026-09-28 14:04 UTCEG score recompute
  21. 2026-09-28 14:04 UTCGHSA enrichment
  22. 2026-09-28 13:50 UTCEPSS rescore
  23. 2026-09-27 14:18 UTCEG score recompute
  24. 2026-09-27 14:18 UTCGHSA enrichment
  25. 2026-09-26 14:34 UTCEG score recompute
Show 2 more
  1. 2026-09-26 13:38 UTCEG score recompute
  2. 2026-09-26 13:30 UTCMITRE cvelistV5first tracked

Related CVEs(same CWE)

Frequently asked(5)

What is CVE-2026-100653?
CVE-2026-100653 is a medium vulnerability published on September 26, 2026. vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the…
When was CVE-2026-100653 disclosed?
CVE-2026-100653 was first published on September 26, 2026, with the most recent update on October 1, 2026. EchelonGraph re-ingests CVE updates from NVD on a 2-hour cycle, so this page reflects the latest published state.
Is CVE-2026-100653 actively exploited?
CVE-2026-100653 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 0.3% probability of exploitation in the next 30 days (19th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-100653?
CVE-2026-100653 has a CVSS base score of 6.5 (the CNA's own assessment, by vulncheck; NVD's own analysis pending).
How do I remediate CVE-2026-100653?
A fix for CVE-2026-100653 is available: update to the fixed version the vendor names in its advisory. The vendor advisories EchelonGraph has for CVE-2026-100653 are linked in the Vendor Advisories panel on this page.

Dependency Blast Radius

Explore the affected products and dependency analysis for CVE-2026-100653

Explore →

Is Your Infrastructure Affected by CVE-2026-100653?

EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.

Book a Demo →Browse All CVEs