CVE-2026-31253

HIGHNVD 7.37.3
EchelonGraph scoreMEDIUM confidence

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

Score 7.3 from GitHub Security Advisory (severity: HIGH) published 2026-05-11. NVD baseline CVSS 7.3; sources differ by 0.0.

Triggered by: GitHub Security Advisory CVSS
Sources: EPSS, GitHub Security Advisory, NVD CVSS v3
7.3EG
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: 0.4%CVSS: 7.3Exploit: None knownExposed services: Not assessed

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

The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 (2025-13-04) contains an insecure deserialization vulnerability (CWE-502) in its checkpoint loading mechanism. The load_checkpoint() function in checkpoint.py and the checkpoint loading code in eval.py use torch.load() without enabling the security-restrictive weights_only=True parameter. This allows the deserialization of arbitrary Python objects via the pickle module. An attacker can exploit this by providing a maliciously crafted checkpoint file. When a victim loads this checkpoint during model warmstarting or evaluation, arbitrary code is executed on the victim's system.

CVSS v3
7.3
EG Score
7.3HIGHmedium confidence
EG Risk
33
EG Risk 33/100

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)
Severity73% × 45%
The EG Score, 7.3 of 10.
Exploitation0.4% × 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
0.4%
EPSS %ILE
29th
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

May 11, 2026

Last Modified

May 12, 2026

Advisory Details (2)

Auto-updated Jun 11, 2026
No fix on record yet.
Web page

Notion

https://www.notion.so/CVE-2026-31253-35d1e1393188813f9e77e2038104bc49
Web page

GitHub - Dao-AILab/flash-attention: Fast and memory-efficient exact attention · GitHub

https://github.com/Dao-AILab/flash-attention

Vendor Advisories for CVE-2026-31253(1)

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

Patch Availability(1)

Vendor / EcosystemFix (by version range) / PatchReleasedSource
pipflash_attn—ghsa

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
flash-attn0.2.0 ... 2.8.3 (75 versions)
  • every version through 2.8.3: no fix on record
—

Weakness Classification(2)

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

Data Freshness Timeline

(refreshed 11× in last 7d / 27× 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 255 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 159 oldest rows on record are not shown and were not checked, so the counts can still include repeats.

  1. 2026-10-15 16:31 UTCEPSS rescoreEPSS 0.4% · 29th percentile
  2. 2026-10-14 16:32 UTCEPSS rescoreEPSS 0.4% · 29th percentile
  3. 2026-10-13 16:34 UTCEPSS rescoreEPSS 0.4% · 29th percentile
  4. 2026-10-12 16:34 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  5. 2026-10-12 16:15 UTCEG score recompute
  6. 2026-10-11 19:57 UTCEG score recompute
  7. 2026-10-11 17:11 UTCEG score recompute
  8. 2026-10-11 14:39 UTCEG score recompute
  9. 2026-10-11 13:48 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  10. 2026-10-10 20:00 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  11. 2026-10-09 20:07 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  12. 2026-10-08 17:12 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  13. 2026-10-07 23:47 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  14. 2026-10-06 15:17 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  15. 2026-10-05 17:42 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  16. 2026-10-04 23:22 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  17. 2026-10-03 14:25 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  18. 2026-10-01 19:49 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  19. 2026-09-30 15:03 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  20. 2026-09-28 13:51 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  21. 2026-09-26 15:58 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  22. 2026-09-26 00:26 UTCEPSS rescoreEPSS 0.4% · 28th percentile
  23. 2026-09-20 20:15 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  24. 2026-09-18 19:27 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  25. 2026-09-17 19:30 UTCEPSS rescoreEPSS 0.2% · 12th percentile
Show 71 more
  1. 2026-09-16 14:08 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  2. 2026-09-16 05:15 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  3. 2026-09-13 16:47 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  4. 2026-09-12 15:01 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  5. 2026-09-11 14:52 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  6. 2026-09-10 09:34 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  7. 2026-09-08 22:00 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  8. 2026-09-07 16:01 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  9. 2026-09-06 13:47 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  10. 2026-09-04 05:06 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  11. 2026-09-02 14:11 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  12. 2026-09-01 13:53 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  13. 2026-08-30 19:17 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  14. 2026-08-30 01:22 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  15. 2026-08-28 21:41 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  16. 2026-08-27 14:25 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  17. 2026-08-26 14:46 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  18. 2026-08-25 13:49 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  19. 2026-08-24 14:17 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  20. 2026-08-23 18:39 UTCEG score recompute
  21. 2026-08-23 18:39 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  22. 2026-08-23 00:19 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  23. 2026-08-21 23:49 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  24. 2026-08-20 22:55 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  25. 2026-08-19 17:03 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  26. 2026-08-18 13:48 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  27. 2026-08-17 13:47 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  28. 2026-08-17 11:11 UTCEG score recompute
  29. 2026-08-17 11:11 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  30. 2026-08-16 14:56 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  31. 2026-08-16 04:26 UTCEG score recompute
  32. 2026-08-16 04:26 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  33. 2026-08-16 02:14 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  34. 2026-08-15 01:30 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  35. 2026-08-14 15:25 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  36. 2026-08-14 02:59 UTCEG score recompute
  37. 2026-08-14 02:59 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  38. 2026-08-13 22:00 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  39. 2026-08-13 04:56 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  40. 2026-08-12 16:31 UTCEG score recompute
  41. 2026-08-12 16:31 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  42. 2026-08-12 13:50 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  43. 2026-08-11 15:47 UTCEG score recompute
  44. 2026-08-11 15:47 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  45. 2026-08-11 01:11 UTCEG score recompute
  46. 2026-08-11 01:11 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  47. 2026-08-11 00:00 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  48. 2026-08-10 03:06 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  49. 2026-08-09 14:41 UTCEG score recompute
  50. 2026-08-09 14:41 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  51. 2026-08-09 13:46 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  52. 2026-08-09 02:15 UTCEG score recompute
  53. 2026-08-09 02:15 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  54. 2026-08-08 16:37 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  55. 2026-08-08 02:30 UTCEG score recompute
  56. 2026-08-08 02:30 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  57. 2026-08-07 14:04 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  58. 2026-08-07 01:38 UTCEG score recompute
  59. 2026-08-07 01:38 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  60. 2026-08-06 13:46 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  61. 2026-08-06 10:24 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  62. 2026-08-05 21:58 UTCEG score recompute
  63. 2026-08-05 21:58 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  64. 2026-08-05 19:17 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  65. 2026-08-04 15:10 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  66. 2026-08-04 10:38 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  67. 2026-08-03 19:26 UTCEG score recompute
  68. 2026-08-03 19:26 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  69. 2026-08-03 10:36 UTCEPSS rescoreEPSS 0.2% · 12th percentile
  70. 2026-08-03 03:02 UTCGHSA enrichmentGHSA-7g5w-pq96-8c5w · GitHub severity HIGH · GitHub CVSS 7.3
  71. 2026-08-02 14:36 UTCEG score recompute

Related CVEs(same CWE)

Frequently asked(5)

What is CVE-2026-31253?
CVE-2026-31253 is a high-severity vulnerability; its CVE record was published on May 11, 2026. The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 (2025-13-04) contains an insecure deserialization vulnerability (CWE-502) in its checkpoint loading mechanism. The loadcheckpoint() function in checkpoint.py and the checkpoint loading code in eval.py use…
When was the CVE record for CVE-2026-31253 published?
The CVE record for CVE-2026-31253 was published on May 11, 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 May 12, 2026; the record at its source may have been updated since.
Is CVE-2026-31253 actively exploited?
CVE-2026-31253 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 0.4% probability of exploitation in the next 30 days (29th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-31253?
CVE-2026-31253 has a CVSS v3 base score of 7.3 (NVD).
How do I remediate CVE-2026-31253?
No fix for CVE-2026-31253 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-2026-31253 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-2026-31253

Explore →

Is Your Infrastructure Affected by CVE-2026-31253?

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

Book a Demo →Browse All CVEs