CVE-2026-90554

HIGHNVD 7.57.5—
EchelonGraph scoreMEDIUM confidence

This high-severity CVE scores 7.5 under NVD CVSS v3. EPSS exploit probability: 0.2% (9th percentile of EPSS-scored CVEs). GitHub Security Advisory CVSS 6.2 is lower than this score; a lower GitHub score revises it only when its advisory is newer than the CVE record's last change, and this advisory (published 2026-09-12) is not newer than that change (2026-09-24), so it does not.

Triggered by: Headline CVSS from NVD CVSS v3
Sources: EPSS, GitHub Security Advisory, NVD CVSS v3
Trending — 3 sources updated this week
7.5EG
EchelonGraph verdictPlan mitigationRemediate within your standard update timelines (the SSVC decision on this page).
  • High severity, but no confirmed exploitation yet
CISA-KEV: Not listedEPSS PROB: 0.2%CVSS: 7.5Exploit: None knownExposed services: Not assessed

No fix is confirmed yet — apply a workaround or compensating control (WAF / firewall / segmentation) and watch for the fix.

vLLM versions >=0.10.2 and <0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(video_bytes)) without the max_duration_s or max_decode_bytes parameters, so neither VLLM_MAX_AUDIO_DECODE_DURATION_S nor VLLM_MAX_AUDIO_DECODE_BYTES is enforced (unlike the direct audio upload path in AudioMediaIO). When a NanoNemotronVL model is served with use_audio_in_video=True, an attacker who supplies a small, highly compressed video as multimodal input can force the server to allocate gigabytes of memory during audio decoding, resulting in a denial of service. Fixed in vLLM 0.28.0.

CVSS v3
7.5
EG Score
7.5HIGHmedium confidence
EG Risk
34
EG Risk 34/100

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 (value × EchelonGraph’s weight)
Severity75% × 45%
The EG Score, 7.5 of 10.
Exploitation0.2% × 40%
FIRST EPSS: the 30-day exploitation probability when EchelonGraph scored this CVE.
Automatability0% × 15%
CISA's SSVC Automatable value: no (Vulnrichment).
SSVC (CISA's decision table, applied by EchelonGraph): Track at every mission impact level.
Exploitation none (CISA Vulnrichment) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE.
Action: No fix is confirmed yet. Restrict network exposure of the affected system or apply the vendor's mitigation within your standard update timelines, and watch the vendor's advisory for the fix.
EPSS PROB
0.2%
EPSS %ILE
9th
KEV
Not listed

SSVC (CISA's decision table, applied by EchelonGraph)Track at every mission impact level.

No fix is confirmed yet. Restrict network exposure of the affected system or apply the vendor's mitigation within your standard update timelines, and watch the vendor's advisory for the fix.

Exploitation none (CISA Vulnrichment) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE. Mission impact is CISA's Mission & Well-being decision point, and only you can judge it. CISA's table makes it high when Mission Prevalence is Essential — the vulnerable component "directly provides capabilities that constitute at least one MEF for at least one entity" (MEF: mission essential function) — or when Public Well-Being Impact is Irreversible: "multiple fatalities are likely", the cyber-physical system "is likely lost or destroyed", "extreme or serious externalities" are imposed on other parties, or social systems such as elections or the financial grid "are destabilized and potentially collapse". CISA's decision table

BOD 26-04 (CISA's remediation timeline, Table 1)60 days if the affected system is publicly exposed; Fix on system upgrade if it is not.

In CISA KEV no (CISA's KEV catalog) · Automatable no (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). Whether your system is publicly exposed is yours to answer. BOD 26-04 binds US Federal Civilian Executive Branch agencies; for anyone else it is CISA's published timeline, a reference rather than an obligation. The directive · how EchelonGraph applies it

Published

September 12, 2026

Last Modified

September 24, 2026

Advisory Details (2)

Auto-updated Sep 12, 2026
No fix on record yet. Sources: github.
Web page

vLLM before 0.28.0 Denial of Service via audio extraction | Advisories | VulnCheck

https://www.vulncheck.com/advisories/vllm-before-0.28.0-denial-of-service-via-audio-extraction
GitHub advisoryMentions a patch (unverified)

Denial of Service NanoNemoTronVL Video Audio Extraction Bomb · Advisory · vllm-project/vllm · GitHub

https://github.com/vllm-project/vllm/security/advisories/GHSA-936p-m5pv-vvjf

Vendor Advisories for CVE-2026-90554(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 31× in last 7d / 105× in last 30d)

Each row is a time one of our pipelines fetched, updated or re-scored this CVE. 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. An arrow marks a change the record states ("CVSS v3 → 9.1"); EPSS and GHSA rows show the values recorded at that time, not what changed (EPSS values are rounded; hover one for the exact figures). Most recent first.

Showing the most recent 99 of 105 refreshes on record for this CVE.

A repeat, a row identical to one the same pipeline wrote less than 10 seconds earlier, is not a separate refresh and is not counted. 1 repeat row among the latest 100 is merged into the row it repeats. The 6 oldest rows on record are not shown and were not checked, so the counts can still include repeats.

  1. 2026-10-11 19:54 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  2. 2026-10-11 19:14 UTCEG score recompute
  3. 2026-10-11 16:30 UTCEG score recompute
  4. 2026-10-11 13:50 UTCEPSS rescoreEPSS 0.2% · 9th percentile
  5. 2026-10-11 13:45 UTCEG score recompute
  6. 2026-10-11 06:40 UTCEG score recompute
  7. 2026-10-11 06:40 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  8. 2026-10-10 20:00 UTCEPSS rescoreEPSS 0.2% · 9th percentile
  9. 2026-10-10 17:12 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  10. 2026-10-10 04:53 UTCEG score recompute
  11. 2026-10-10 04:53 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  12. 2026-10-09 20:08 UTCEPSS rescoreEPSS 0.2% · 9th percentile
  13. 2026-10-09 16:21 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  14. 2026-10-08 23:08 UTCEG score recompute
  15. 2026-10-08 23:08 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  16. 2026-10-08 17:14 UTCEPSS rescoreEPSS 0.2% · 9th percentile
  17. 2026-10-08 07:34 UTCEG score recompute
  18. 2026-10-08 07:34 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  19. 2026-10-07 23:48 UTCEPSS rescoreEPSS 0.2% · 9th percentile
  20. 2026-10-07 19:17 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  21. 2026-10-07 06:51 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  22. 2026-10-06 18:34 UTCEG score recompute
  23. 2026-10-06 18:34 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  24. 2026-10-06 15:18 UTCEPSS rescoreEPSS 0.2% · 9th percentile
  25. 2026-10-06 06:17 UTCEG score recompute
Show 74 more
  1. 2026-10-06 06:17 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  2. 2026-10-05 17:44 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  3. 2026-10-05 17:32 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  4. 2026-10-05 05:15 UTCEG score recompute
  5. 2026-10-05 05:15 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  6. 2026-10-04 23:22 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  7. 2026-10-04 16:58 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  8. 2026-10-04 04:34 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  9. 2026-10-03 16:18 UTCEG score recompute
  10. 2026-10-03 16:18 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  11. 2026-10-03 04:02 UTCEG score recompute
  12. 2026-10-03 04:02 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  13. 2026-10-02 15:39 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  14. 2026-10-02 03:23 UTCEG score recompute
  15. 2026-10-02 03:23 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  16. 2026-10-01 19:51 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  17. 2026-10-01 14:26 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  18. 2026-10-01 02:10 UTCEG score recompute
  19. 2026-10-01 02:10 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  20. 2026-09-30 15:04 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  21. 2026-09-30 13:54 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  22. 2026-09-30 01:36 UTCEG score recompute
  23. 2026-09-30 01:36 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  24. 2026-09-29 11:26 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  25. 2026-09-28 22:59 UTCEG score recompute
  26. 2026-09-28 22:59 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  27. 2026-09-28 13:52 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  28. 2026-09-28 10:42 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  29. 2026-09-27 22:15 UTCEG score recompute
  30. 2026-09-27 22:15 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  31. 2026-09-27 13:49 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  32. 2026-09-27 09:58 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  33. 2026-09-26 21:42 UTCEG score recompute
  34. 2026-09-26 21:42 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  35. 2026-09-26 15:59 UTCEPSS rescoreEPSS 0.2% · 8th percentile
  36. 2026-09-26 09:25 UTCEG score recompute
  37. 2026-09-26 09:25 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  38. 2026-09-25 21:09 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  39. 2026-09-25 08:51 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  40. 2026-09-24 20:35 UTCEG score recompute▲ 1.30
  41. 2026-09-24 20:35 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  42. 2026-09-24 20:35 UTCNVD updateCVSS v3 → 7.5 · severity → HIGH
  43. 2026-09-24 16:03 UTCEG score recompute
  44. 2026-09-24 16:03 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  45. 2026-09-24 14:04 UTCEPSS rescoreEPSS 0.2% · 4th percentile
  46. 2026-09-23 17:54 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  47. 2026-09-23 16:36 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  48. 2026-09-22 17:10 UTCEG score recompute
  49. 2026-09-22 17:10 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  50. 2026-09-22 16:01 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  51. 2026-09-21 21:09 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  52. 2026-09-21 17:35 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  53. 2026-09-20 18:04 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  54. 2026-09-19 18:36 UTCEG score recompute
  55. 2026-09-19 18:36 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  56. 2026-09-18 19:28 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  57. 2026-09-18 19:10 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  58. 2026-09-17 19:41 UTCEG score recompute
  59. 2026-09-17 19:41 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  60. 2026-09-17 19:32 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  61. 2026-09-16 20:14 UTCEG score recompute
  62. 2026-09-16 20:14 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  63. 2026-09-16 14:09 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  64. 2026-09-16 05:15 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  65. 2026-09-15 20:48 UTCEG score recompute
  66. 2026-09-15 20:48 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  67. 2026-09-15 03:12 UTCEPSS rescoreEPSS 0.1% · 1st percentile
  68. 2026-09-14 21:17 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  69. 2026-09-14 19:21 UTCEG score recompute
  70. 2026-09-14 19:21 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  71. 2026-09-14 18:28 UTCEG score recompute
  72. 2026-09-14 18:28 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2
  73. 2026-09-14 12:17 UTCEG score recompute
  74. 2026-09-14 12:17 UTCGHSA enrichmentGHSA-m39m-2pqx-3vm2 · GitHub CVSS 6.2

Related CVEs(same CWE)

Frequently asked(5)

What is CVE-2026-90554?
CVE-2026-90554 is a high-severity vulnerability; its CVE record was published on September 12, 2026. vLLM versions >=0.10.2 and <0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nanonemotronvl.py, extractaudiofromvideos calls loadaudiopyav(BytesIO(videobytes)) without the maxdurations or maxdecodebytes parameters, so…
When was CVE-2026-90554 made public?
CVE-2026-90554 was made public on August 28, 2026: that is the date its CVE record's CNA, VulnCheck, gives for when the vulnerability was made public. The CVE record itself was published on September 12, 2026. EchelonGraph's copy of the record carries a last-modified date of September 24, 2026; the record at its source may have been updated since.
Is CVE-2026-90554 actively exploited?
CVE-2026-90554 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 0.2% probability of exploitation in the next 30 days (9th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-90554?
CVE-2026-90554 has a CVSS v3 base score of 7.5 (NVD).
How do I remediate CVE-2026-90554?
No fix for CVE-2026-90554 is confirmed yet. Until one is published, restrict network exposure of the affected system or apply the vendor's mitigation — for example, keep it off the internet or limit it to trusted networks — and watch the vendor's advisory for the fix. The vendor advisories EchelonGraph has for CVE-2026-90554 are linked in the Vendor Advisories panel on this page.

Dependency Blast Radius

Explore the affected products and dependency analysis for CVE-2026-90554

Explore →

Is Your Infrastructure Affected by CVE-2026-90554?

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

Book a Demo →Browse All CVEs