CVE-2026-73558

MEDIUMPre-NVD 5.35.3
EchelonGraph scoreMEDIUM confidence

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

This medium-severity CVE scores 5.3 under a secondary CVSS source (NVD's own analysis pending). EPSS exploit probability: 0.4% (33rd 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
Elevated
5.3EG
EchelonGraph verdictPlan a fixRemediate within your standard update timelines (the SSVC decision on this page).
  • CISA's Vulnrichment rates exploitation as PoC; EchelonGraph holds no exploit-code reference for it
CISA-KEV: Not listedEPSS PROB: 0.4%CVSS: 5.3Exploit: No code on recordExposed services: Not assessed

A fix is available — apply it.

vLLM is an inference and serving engine for large language models. Prior to 0.27.0, an integer overflow in blockIdx.x * 2 * d in activation_kernels.cu can cause act_and_mul_kernel to consume another batched user's input, allowing a request processed in the same inference batch to receive a partial or complete copy of another user's inference result. This issue is fixed in version 0.27.0.

CVSS v3
5.3
EG Score
5.3MEDIUMmedium confidence
EG Risk
40
EG Risk 40/100

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)
Severity53% × 45%
The EG Score, 5.3 of 10.
Exploitation40% × 40%
EchelonGraph's fixed weight for EG-KEV Tier 3 (Elevated): a constant, not a probability.
Automatability0% × 15%
CISA's SSVC Automatable value: no (Vulnrichment).
SSVC (CISA's decision table, applied by EchelonGraph): Track at low or medium mission impact; Track* at high (mission-essential systems).
Exploitation public PoC (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: A fix is available. Apply it within your standard update timelines.
EPSS PROB
0.4%
EPSS %ILE
33rd
KEV
Not listed

SSVC (CISA's decision table, applied by EchelonGraph)Track 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). 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

August 13, 2026

Last Modified

September 28, 2026

Advisory Details (5)

Auto-updated Aug 13, 2026
Patch available. Sources: github_commit, github_pr, github_release, github.
GitHub advisoryMentions a patch (unverified)

Cross-User Data Leak Vulnerability · Advisory · vllm-project/vllm · GitHub

https://github.com/vllm-project/vllm/security/advisories/GHSA-7m6h-x95x-82q5
GitHub release Patch Available

v0.27.0

Patch available: vllm-project/vllm v0.27.0

https://github.com/vllm-project/vllm/releases/tag/v0.27.0
GitHub pull request

[Bugfix][Kernel] Fix integer overflow in libtorch_stable/activation_kernels.cu

Fix merged in vllm-project/vllm PR #49660 on 2026-07-30 — awaiting tagged release

https://github.com/vllm-project/vllm/pull/49660
Web page

[Bug]: integer overflow in activation_kernels.cu · Issue #42860 · vllm-project/vllm · GitHub

https://github.com/vllm-project/vllm/issues/42860
GitHub commit

commit 451227cb3ff0 (vllm-project/vllm)

Fix landed in vllm-project/vllm commit 451227cb3ff0 — awaiting tagged release

https://github.com/vllm-project/vllm/commit/451227cb3ff07989698fed982c2d3e4300257924

Affected Packages

(1 across 1 ecosystem)
PyPI(1)
PackageVulnerable rangeFix by version rangeDependents
vllm0.0.1 ... 0.9.2 (93 versions)
  • every version up to 0.27.0: fixed in 0.27.0
—

Weakness Classification(1)

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

Data Freshness Timeline

(refreshed 11× in last 7d / 45× 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 96 of 101 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. 4 repeat rows among the latest 100 are merged into the row they repeat. The 5 oldest rows on record are not shown and were not checked, so the counts can still include repeats.

  1. 2026-10-13 16:35 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  2. 2026-10-12 16:34 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  3. 2026-10-12 15:40 UTCEG score recompute
  4. 2026-10-11 19:26 UTCEG score recompute
  5. 2026-10-11 16:42 UTCEG score recompute
  6. 2026-10-11 14:01 UTCEG score recompute
  7. 2026-10-11 13:49 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  8. 2026-10-10 20:00 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  9. 2026-10-09 20:07 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  10. 2026-10-08 17:13 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  11. 2026-10-07 23:48 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  12. 2026-10-06 19:28 UTCEG score recompute
  13. 2026-10-06 15:18 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  14. 2026-10-05 20:46 UTCEG score recompute
  15. 2026-10-05 17:43 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  16. 2026-10-04 23:22 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  17. 2026-10-03 19:48 UTCEG score recompute
  18. 2026-10-03 14:26 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  19. 2026-10-02 07:33 UTCEG score recompute
  20. 2026-10-01 19:50 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  21. 2026-09-30 21:47 UTCEG score recompute
  22. 2026-09-29 15:24 UTCEPSS rescoreEPSS 0.4% · 33rd percentile
  23. 2026-09-28 18:56 UTCEG score recompute
  24. 2026-09-28 13:52 UTCEPSS rescoreEPSS 0.4% · 32nd percentile
  25. 2026-09-28 08:11 UTCEG score recompute
Show 71 more
  1. 2026-09-27 13:48 UTCEPSS rescoreEPSS 0.4% · 32nd percentile
  2. 2026-09-27 09:44 UTCEG score recompute
  3. 2026-09-26 15:59 UTCEPSS rescoreEPSS 0.4% · 31st percentile
  4. 2026-09-25 03:33 UTCEG score recompute
  5. 2026-09-23 01:16 UTCEG score recompute
  6. 2026-09-22 02:48 UTCEG score recompute
  7. 2026-09-21 21:09 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  8. 2026-09-21 04:16 UTCEG score recompute
  9. 2026-09-20 20:16 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  10. 2026-09-20 05:31 UTCEG score recompute
  11. 2026-09-19 07:04 UTCEG score recompute
  12. 2026-09-18 19:28 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  13. 2026-09-18 08:36 UTCEG score recompute
  14. 2026-09-17 10:09 UTCEG score recompute
  15. 2026-09-16 14:08 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  16. 2026-09-16 11:41 UTCEG score recompute
  17. 2026-09-16 05:15 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  18. 2026-09-15 13:04 UTCEG score recompute
  19. 2026-09-15 03:12 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  20. 2026-09-14 13:53 UTCEG score recompute
  21. 2026-09-12 16:51 UTCEG score recompute
  22. 2026-09-12 15:02 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  23. 2026-09-11 18:23 UTCEG score recompute
  24. 2026-09-11 14:53 UTCEPSS rescoreEPSS 0.3% · 25th percentile
  25. 2026-09-10 19:55 UTCEG score recompute
  26. 2026-09-10 09:35 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  27. 2026-09-09 04:57 UTCEG score recompute
  28. 2026-09-08 22:01 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  29. 2026-09-07 19:57 UTCEG score recompute
  30. 2026-09-06 21:29 UTCEG score recompute
  31. 2026-09-06 13:48 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  32. 2026-09-05 23:00 UTCEG score recompute
  33. 2026-09-05 00:32 UTCEG score recompute
  34. 2026-09-04 05:07 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  35. 2026-09-03 01:02 UTCEG score recompute
  36. 2026-09-02 14:12 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  37. 2026-09-02 01:49 UTCEG score recompute
  38. 2026-09-01 13:54 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  39. 2026-09-01 04:40 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  40. 2026-08-31 04:34 UTCEG score recompute
  41. 2026-08-30 19:17 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  42. 2026-08-30 06:06 UTCEG score recompute
  43. 2026-08-30 01:22 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  44. 2026-08-29 07:37 UTCEG score recompute
  45. 2026-08-28 21:42 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  46. 2026-08-27 19:32 UTCEG score recompute
  47. 2026-08-27 14:25 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  48. 2026-08-26 19:07 UTCEG score recompute
  49. 2026-08-26 14:47 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  50. 2026-08-25 20:40 UTCEG score recompute
  51. 2026-08-25 13:49 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  52. 2026-08-24 21:58 UTCEG score recompute
  53. 2026-08-23 01:03 UTCEG score recompute
  54. 2026-08-23 00:19 UTCEPSS rescoreEPSS 0.3% · 18th percentile
  55. 2026-08-22 02:35 UTCEG score recompute
  56. 2026-08-21 23:49 UTCEPSS rescoreEPSS 0.3% · 18th percentile
  57. 2026-08-21 04:07 UTCEG score recompute
  58. 2026-08-20 22:56 UTCEPSS rescoreEPSS 0.3% · 18th percentile
  59. 2026-08-20 05:40 UTCEG score recompute
  60. 2026-08-19 17:04 UTCEPSS rescoreEPSS 0.3% · 18th percentile
  61. 2026-08-19 07:12 UTCEG score recompute
  62. 2026-08-18 13:49 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  63. 2026-08-18 08:42 UTCEG score recompute
  64. 2026-08-17 13:47 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  65. 2026-08-17 10:14 UTCEG score recompute
  66. 2026-08-16 14:56 UTCEPSS rescoreEPSS 0.3% · 18th percentile
  67. 2026-08-16 11:47 UTCEG score recompute
  68. 2026-08-16 02:15 UTCEPSS rescoreEPSS 0.3% · 18th percentile
  69. 2026-08-15 13:20 UTCEG score recompute
  70. 2026-08-15 01:31 UTCEPSS rescoreEPSS 0.3% · 17th percentile
  71. 2026-08-14 14:53 UTCEG score recompute

Frequently asked(5)

What is CVE-2026-73558?
CVE-2026-73558 is a medium-severity vulnerability; its CVE record was published on August 13, 2026. vLLM is an inference and serving engine for large language models. Prior to 0.27.0, an integer overflow in blockIdx.x 2 d in activationkernels.cu can cause actandmulkernel to consume another batched user's input, allowing a request processed in the same inference batch to receive a partial or…
When was the CVE record for CVE-2026-73558 published?
The CVE record for CVE-2026-73558 was published on August 13, 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 September 28, 2026; the record at its source may have been updated since.
Is CVE-2026-73558 actively exploited?
CVE-2026-73558 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 0.4% probability of exploitation in the next 30 days (33rd percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-73558?
CVE-2026-73558 has a CVSS base score of 5.3 (a secondary CVSS source that NVD displays; NVD's own analysis pending).
How do I remediate CVE-2026-73558?
A fix for CVE-2026-73558 is available: update to the fixed version the vendor names in its advisory. The vendor advisories EchelonGraph has for CVE-2026-73558 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-2026-73558

Explore →

Is Your Infrastructure Affected by CVE-2026-73558?

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