Skip to content

← Actionable
High, CVSS 7.3amended

flash-attention contains an insecure deserialization vulnerability in its checkpoint loading mechanism

CVE-2026-31253GHSA-7g5w-pq96-8c5w

What to do

Vendor response not established. Follow the primary source's mitigation guidance and check back for a fix.

This page has no inventory attached, so it cannot say whether you are affected. A vendor fix is not proof that your deployment is fixed.

Affected packages

  • PyPI flash-attn

    Affected
    <= 2.8.3
    Fixed in
    no fixed version recorded

Sources

Severity and exploitation
Exploitation
No confirmed exploitation recorded. Patch Next marks exploitation only from CISA KEV.
CVSS
7.3 · CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L
EPSS
  • CVE-2026-31253: 0.22% estimated exploitation activity in the next 30 days; percentile 12.37 (FIRST EPSS, 2026-11-03). This is a model estimate, not proof of exploitation or customer exposure.
Source description

As published by the source; long descriptions are shortened.

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 torch.load() without enabling the security-restrictive weightsonly=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.

Dates and revisions
Published by the source
Last modified by the source
Source last checked
  1. Revision 2 · amended ·

    Amended from ghsa record GHSA-7g5w-pq96-8c5w

  2. Revision 1 · created ·

    Published from osv record GHSA-7g5w-pq96-8c5w

Agent brief

Preview the exact text before you copy it. Commands are read-only suggestions; review them before running anything.

# Patch Next advisory brief: CVE-2026-31253

Advisory: flash-attention contains an insecure deserialization vulnerability in its checkpoint loading mechanism
IDs: CVE-2026-31253, GHSA-7g5w-pq96-8c5w
Record: https://patchnext.com/advisories/cve-2026-31253 (revision 2, source checked 26 Oct 2026, 13:46 UTC)
Inventory: none attached. This brief comes from a public page and does not say whether any deployment is affected.
Severity: High (CVSS 7.3)
Exploitation: no confirmed exploitation recorded. Patch Next confirms exploitation only from CISA KEV.
Vendor response: Vendor response not established. A vendor fix does not mean a deployment is fixed.

## Affected components
- PyPI flash-attn: affected <= 2.8.3

## Suggested inspection (read-only)
- `python -m pip show flash-attn`: prints the installed version. Run with the environment that serves production active.

## Supported action
No complete vendor fix is recorded. Follow the primary source's mitigation guidance and check this record again before closing anything.

## Required regression checks
- Run the test suite and a production build.
- Exercise the features that depend on flash-attn.
- Repeat the inspection and confirm that only fixed versions remain installed.

## Rollback considerations
- Keep the previous lockfile and deployment artifact. Rolling back reinstates the vulnerable version, so apply any documented mitigation while you fix forward.

## Unresolved questions
- Which deployments run an affected version?
- Exploitation status can change. Check again before lowering the priority.
- Mark the work resolved only after the fixed version is verified in production.

## Sources
- Verified: https://osv.dev/vulnerability/GHSA-7g5w-pq96-8c5w (OSV record GHSA-7g5w-pq96-8c5w lists affected ranges: PyPI flash-attn >=0 <=2.8.3)
- Verified: https://github.com/advisories/GHSA-7g5w-pq96-8c5w (GitHub-reviewed advisory GHSA-7g5w-pq96-8c5w lists affected ranges: PyPI flash-attn >=0 <=2.8.3)

Commands are suggestions: review each one before running it.

Loading voting…