Vulnpedia
← All CVEs

CVE-2026-24233

HIGH 8.4 ALL-YEARS

Published 2026-07-14 · Last modified 2026-07-15 · nvidia · NVD: Deferred

NVIDIA TensorRT-LLM for Linux contains a vulnerability in the restricted unpickler used for model weight deserialization, where a local, unauthenticated attacker could cause deserialization of untrusted data. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, data tampering, and information disclosure.

ELEVATED IMPACT

Severe if exploited (CVSS 8.4), but no confirmed exploitation. Patch on a normal cadence.

Exploitation likelihood

0.3%chance of exploitation in 30 days · 18th percentile

○ In CISA KEV○ Public exploit / PoC◆ SSVC: exploitation none, automatable no

Impact if exploited

8.4CVSS 3.1 · HIGH · CNA

  • ConfidentialityHigh
  • IntegrityHigh
  • AvailabilityHigh

What an attacker needs

  • ⚠Access: Requires local access to the host
  • ✓Privileges: No account or privileges required
  • ✓User interaction: No user interaction needed
  • ✓Complexity: No special conditions — reliably repeatable

✓ lowers the bar for an attacker · ⚠ raises it

Proof of concept & exploit code

No public exploit or proof-of-concept code is catalogued for this CVE (no ExploitDB, Nuclei, or CISA-KEV entry). It may still exist — search GitHub or check the references below.

Affected

Vendors Nvidia

Products Tensorrt Llm

Weakness (CWE)

  • CWE-502: Deserialization of Untrusted Data

CVSS vector

CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

All CVSS metrics

  • HIGH 8.4 v3.1 · CNA Primary
    CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
  • HIGH 8.4 v3.1 · NVD Secondary
    CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

References

Technical & other

Sources: NVD · CVE.org · EPSS