Skip to main content
VulnerabilityPublished

CVE-2026-116840: ggml-org llama.cpp · Integer Overflow or Wraparound

Affects ggml-org llama.cppSource CVE Database V5

0

Mentioned vulnerabilities

Not known to be exploited · CVSS 6.9 Medium : upgrade to b11275 or 0.6.0 when convenient.

Data as of 27 Oct 2026, 01:27 UTC

CVSS 4.0 base score
6.9Medium
How severe the flaw is on the 0–10 CVSS scale.
Patch
Available – fixed in b11275, 0.6.0
CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:L/VI:N/VA:H/SC:N/SI:N/SA:N

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 27 Oct 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponised: no evidence on record.

    No public exploit code indexed in Metasploit or Exploit-DB

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 26 Oct 2026.

What to do

Upgrade llama.cpp to b11275 or 0.6.0; llama.cpp to 0.6.0.

Affected software

ProductAffectedFixed in
ggml-orgllama.cppAffected
  • >=0.1.0 <0.6.0
  • =b5854
Fixed in
  • b11275
  • 0.6.0
GitHubllama.cpppkg:github/llama.cppOther threatsAffected
  • >=0.1.0 <0.6.0
  • =b5854
Fixed in
  • 0.6.0

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

A fix is available in llama.cpp versions b11275 and 0.6.0. Users should upgrade to these or later versions to remediate the vulnerability. No other mitigation steps are indicated.

Description

An integer overflow vulnerability exists in llama.cpp versions from b5854 before b11275 (including versions >=0.1.0 <0.6.0). The flaw occurs in the gguf_init_from_reader function, allowing crafted GGUF files to bypass size checks by padding tensor sizes near the maximum integer value. This results in an undersized data blob and out-of-bounds reads by consumers.

References

Community reviews

Loading reviews…