CVE-2026-22778

CRITICALPre-NVD 9.89.8
EchelonGraph scoreMEDIUM confidence

“—” beside the score: it is equal to the CVSS score beside it.

This critical-severity CVE scores 9.8 under a secondary CVSS source (NVD's own analysis pending). EPSS exploit probability: 17% (96th percentile of EPSS-scored CVEs). No GitHub Security Advisory CVSS is on record for this CVE.

Triggered by: Headline CVSS from a secondary CVSS source
Sources: EPSS, a secondary CVSS source
Trending — 4 sources updated this weekWeaponized
9.8EG
EchelonGraph verdictPlan a fixRemediate within your standard update timelines at low or medium mission impact and sooner than that at high (the SSVC decision on this page).
  • Public exploit code is available (Metasploit)
CISA-KEV: Not listedEPSS PROB: 17%CVSS: 9.8Exploit: MetasploitExposed services: Not assessed

A fix is available — apply it.

vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion guesses to ~8 guesses. This vulnerability can be chained a heap overflow with JPEG2000 decoder in OpenCV/FFmpeg to achieve remote code execution. This vulnerability is fixed in 0.14.1.

CVSS v3
9.8
EG Score
9.8CRITICALmedium confidence
EG Risk
85
EG Risk 85/100

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)
Severity98% × 45%
The EG Score, 9.8 of 10.
Exploitation65% × 40%
EchelonGraph's fixed weight for EG-KEV Tier 2 (Weaponized): a constant, not a probability.
Automatability100% × 15%
CISA's SSVC Automatable value: yes (Vulnrichment).
SSVC (CISA's decision table, applied by EchelonGraph): Track at low or medium mission impact; Attend at high (mission-essential systems).
Exploitation none (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact total (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: A fix is available. Apply it within your standard update timelines at low or medium mission impact and sooner than that at high.
EPSS PROB
17%
EPSS %ILE
96th
KEV
Not listed

SSVC (CISA's decision table, applied by EchelonGraph)Track at low or medium mission impact; Attend at high (mission-essential systems).

A fix is available. Apply it within your standard update timelines at low or medium mission impact and sooner than that at high.

Exploitation none (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact total (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)3 days if the affected system is publicly exposed; 60 days if it is not.

In CISA KEV no (CISA's KEV catalog) · Automatable yes (CISA Vulnrichment) · Technical impact total (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

February 2, 2026

Last Modified

July 15, 2026

Advisory Details (10)

Auto-updated Oct 10, 2026
Patch available. Sources: github_pr, github_release, github, redhat.
Red HatMentions a patch (unverified)
https://access.redhat.com/errata/RHSA-2026:3462
Red HatMentions a patch (unverified)
https://access.redhat.com/errata/RHSA-2026:3461
Red HatMentions a patch (unverified)
https://access.redhat.com/errata/RHSA-2026:30089
Red HatMentions a patch (unverified)
https://access.redhat.com/errata/RHSA-2026:30088
Red HatMentions a patch (unverified)
https://access.redhat.com/errata/RHSA-2026:30087
Red HatMentions a patch (unverified)
https://access.redhat.com/errata/RHSA-2026:19712
GitHub advisoryMentions a patch (unverified)

vLLM RCE In Video Processing · Advisory · vllm-project/vllm · GitHub

https://github.com/vllm-project/vllm/security/advisories/GHSA-4r2x-xpjr-7cvv
GitHub release Patch Available

v0.14.1

Patch available: vllm-project/vllm v0.14.1

https://github.com/vllm-project/vllm/releases/tag/v0.14.1
GitHub pull request

[Frontend] Standardize use of `create_error_response`

Fix merged in vllm-project/vllm PR #32319 on 2026-01-14 — awaiting tagged release

https://github.com/vllm-project/vllm/pull/32319
GitHub pull request

[Frontend] Improve error message

Fix merged in vllm-project/vllm PR #31987 on 2026-01-08 — awaiting tagged release

https://github.com/vllm-project/vllm/pull/31987

Vendor Advisories for CVE-2026-22778(9)

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

Patch Availability(8)

Vendor / EcosystemFix (by version range) / PatchReleasedSource
redhatrhaiis/vllm-rocm-rhel9:17823530932026-06-25redhat
redhatrhaiis/vllm-cuda-rhel9:17823528472026-06-25redhat
redhatrhaiis/vllm-spyre-rhel9:17823529192026-06-25redhat
redhatrhoai/odh-vllm-cpu-rhel9:17782643632026-05-20redhat
redhatrhoai/odh-vllm-gaudi-rhel9:17709560342026-03-04redhat
redhatrhoai/odh-vllm-rocm-rhel9:17720932372026-03-04redhat
redhatrhaiis/vllm-cuda-rhel9:17721605932026-02-27redhat
redhatrhaiis/vllm-rocm-rhel9:17721606252026-02-27redhat

Patches are aggregated from vendor advisories (Red Hat, Microsoft, Cisco, GitHub) and package ecosystems (OSV, GHSA). Multiple rows for the same upstream release have been deduplicated.

Affected Packages

(1 across 1 ecosystem)
PyPI(1)
PackageVulnerable rangeFix by version rangeDependents
vllm0.10.0 ... 0.9.2 (18 versions)
  • 0.8.3 up to 0.14.1: fixed in 0.14.1
—

Weakness Classification(2)

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

Data Freshness Timeline

(refreshed 19× in last 7d / 35× 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 203 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 104 oldest rows on record are not shown and were not checked, so the counts can still include repeats.

  1. 2026-10-15 17:39 UTCEG score recompute
  2. 2026-10-15 17:39 UTCVendor advisory
  3. 2026-10-15 16:31 UTCEPSS rescoreEPSS 17% · 96th percentile
  4. 2026-10-14 16:32 UTCEPSS rescoreEPSS 17% · 96th percentile
  5. 2026-10-14 11:32 UTCEG score recompute
  6. 2026-10-14 11:32 UTCVendor advisory
  7. 2026-10-13 16:33 UTCEPSS rescoreEPSS 17% · 96th percentile
  8. 2026-10-13 10:24 UTCEG score recompute
  9. 2026-10-13 10:24 UTCVendor advisory
  10. 2026-10-12 16:33 UTCEPSS rescoreEPSS 17% · 96th percentile
  11. 2026-10-12 16:22 UTCEG score recompute
  12. 2026-10-12 09:18 UTCVendor advisory
  13. 2026-10-11 20:03 UTCEG score recompute
  14. 2026-10-11 17:17 UTCEG score recompute
  15. 2026-10-11 14:48 UTCEG score recompute
  16. 2026-10-11 13:48 UTCEPSS rescoreEPSS 11% · 95th percentile
  17. 2026-10-10 20:00 UTCEPSS rescoreEPSS 11% · 95th percentile
  18. 2026-10-09 20:07 UTCEPSS rescoreEPSS 11% · 95th percentile
  19. 2026-10-09 09:55 UTCOSV refresh
  20. 2026-10-08 17:11 UTCEPSS rescoreEPSS 11% · 95th percentile
  21. 2026-10-07 23:47 UTCEPSS rescoreEPSS 11% · 95th percentile
  22. 2026-10-06 15:17 UTCEPSS rescoreEPSS 11% · 95th percentile
  23. 2026-10-05 17:42 UTCEPSS rescoreEPSS 11% · 95th percentile
  24. 2026-10-01 19:49 UTCEPSS rescoreEPSS 11% · 95th percentile
  25. 2026-09-30 15:03 UTCEPSS rescoreEPSS 11% · 95th percentile
Show 74 more
  1. 2026-09-29 15:21 UTCEPSS rescoreEPSS 10% · 95th percentile
  2. 2026-09-28 13:51 UTCEPSS rescoreEPSS 10% · 95th percentile
  3. 2026-09-27 13:47 UTCEPSS rescoreEPSS 10% · 95th percentile
  4. 2026-09-26 15:58 UTCEPSS rescoreEPSS 3.8% · 89th percentile
  5. 2026-09-26 00:26 UTCEPSS rescoreEPSS 3.8% · 89th percentile
  6. 2026-09-24 14:03 UTCEPSS rescoreEPSS 3.8% · 89th percentile
  7. 2026-09-19 17:00 UTCOSV refresh
  8. 2026-09-18 19:27 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  9. 2026-09-16 14:07 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  10. 2026-09-16 05:15 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  11. 2026-09-12 15:01 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  12. 2026-09-11 14:52 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  13. 2026-09-10 09:34 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  14. 2026-09-08 22:00 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  15. 2026-09-07 16:01 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  16. 2026-09-06 13:47 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  17. 2026-09-04 05:06 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  18. 2026-09-02 14:11 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  19. 2026-09-01 13:53 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  20. 2026-09-01 04:39 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  21. 2026-08-31 20:05 UTCOSV refresh
  22. 2026-08-30 19:17 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  23. 2026-08-30 01:22 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  24. 2026-08-28 21:41 UTCEPSS rescoreEPSS 3.7% · 89th percentile
  25. 2026-08-27 14:25 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  26. 2026-08-26 14:46 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  27. 2026-08-25 13:48 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  28. 2026-08-23 00:19 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  29. 2026-08-21 23:49 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  30. 2026-08-20 22:55 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  31. 2026-08-19 17:03 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  32. 2026-08-18 13:48 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  33. 2026-08-17 13:47 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  34. 2026-08-16 14:56 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  35. 2026-08-16 02:14 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  36. 2026-08-15 01:30 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  37. 2026-08-14 23:01 UTCVendor advisory
  38. 2026-08-14 15:43 UTCVendor advisory
  39. 2026-08-14 11:40 UTCEG score recompute
  40. 2026-08-14 11:40 UTCVendor advisory
  41. 2026-08-13 22:00 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  42. 2026-08-13 03:50 UTCEG score recompute
  43. 2026-08-13 03:50 UTCVendor advisory
  44. 2026-08-12 13:50 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  45. 2026-08-11 13:42 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  46. 2026-08-11 00:00 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  47. 2026-08-10 10:03 UTCVendor advisory
  48. 2026-08-10 00:15 UTCVendor advisory
  49. 2026-08-09 20:12 UTCVendor advisory
  50. 2026-08-09 16:10 UTCEG score recompute
  51. 2026-08-09 16:10 UTCVendor advisory
  52. 2026-08-09 13:46 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  53. 2026-08-08 16:37 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  54. 2026-08-08 07:06 UTCVendor advisory
  55. 2026-08-08 03:03 UTCVendor advisory
  56. 2026-08-07 23:00 UTCEG score recompute
  57. 2026-08-07 23:00 UTCVendor advisory
  58. 2026-08-06 13:46 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  59. 2026-08-06 08:38 UTCVendor advisory
  60. 2026-08-06 04:32 UTCEG score recompute
  61. 2026-08-06 04:32 UTCVendor advisory
  62. 2026-08-05 19:17 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  63. 2026-08-04 15:10 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  64. 2026-08-04 10:38 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  65. 2026-08-03 10:35 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  66. 2026-08-03 02:02 UTCVendor advisory
  67. 2026-08-02 21:40 UTCEG score recompute
  68. 2026-08-02 21:40 UTCVendor advisory
  69. 2026-08-02 02:26 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  70. 2026-08-01 04:15 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  71. 2026-07-31 11:42 UTCOSV refresh
  72. 2026-07-30 01:30 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  73. 2026-07-27 14:13 UTCEPSS rescoreEPSS 3.7% · 88th percentile
  74. 2026-07-26 14:54 UTCEPSS rescoreEPSS 3.7% · 88th percentile

Publicly available exploits

(2 references)

Public exploit code is referenced for this CVE (1 Metasploit module), as is 1 Nuclei detection template. A detection template checks whether a system is affected; it is not necessarily exploit code. Defenders should treat patch urgency accordingly.

  • Metasploitauxiliary/scanner/http/vllm_multimodal_info_leak✓ verified
    Published Sep 26, 2026

    vLLM Multimodal Heap-Address Information Leak Scanner

    Open source ↗
  • Nucleihttp/cves/2026/CVE-2026-22778.yaml
    Published Jun 28, 2026

    vLLM 0.8.3 - 0.14.0 - Information Disclosure

    Open source ↗

Related CVEs(same product + same vendor + same CWE)

Frequently asked(5)

What is CVE-2026-22778?
CVE-2026-22778 is a critical-severity vulnerability; its CVE record was published on February 2, 2026. vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion…
When was the CVE record for CVE-2026-22778 published?
The CVE record for CVE-2026-22778 was published on February 2, 2026. That is the record's publication date; the vulnerability itself may have been made public earlier. EchelonGraph's copy of the record carries a last-modified date of July 15, 2026; the record at its source may have been updated since.
Is CVE-2026-22778 actively exploited?
CVE-2026-22778 is not currently on CISA's Known Exploited Vulnerabilities catalog. EchelonGraph's EG-KEV model places it in Tier 2 (Weaponized), which is not evidence of exploitation: a Metasploit module exists for it. EchelonGraph does not test whether exploit code works. FIRST EPSS estimates a 17% probability of exploitation in the next 30 days (96th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-22778?
CVE-2026-22778 has a CVSS base score of 9.8 (a secondary CVSS source that NVD displays; NVD's own analysis pending).
How do I remediate CVE-2026-22778?
A fix for CVE-2026-22778 is available: update to the fixed version the vendor names in its advisory. The vendor advisories EchelonGraph has for CVE-2026-22778 are linked in the Vendor Advisories panel on this page.

Dependency Blast Radius

See which npm, PyPI, Go, and Maven packages are affected by CVE-2026-22778

Explore →

Is Your Infrastructure Affected by CVE-2026-22778?

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

Book a Demo →Browse All CVEs