CVSS 4.0
8.5 highCVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:NHigh (8.5) in ggml llama.cpp. EPSS 0.2% estimated probability of exploitation activity in the next 30 days.
CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:NCVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:HExplain CVE-2026-43627 for an engineer deciding whether to act. Use https://cve.bz/api/public/cves/CVE-2026-43627?via=ask-ai-copy and https://cve.bz/cve/CVE-2026-43627?via=ask-ai-copy as the source of truth: what is affected and which versions fix it, whether it is in CISA KEV, public exploits, and its EPSS (an internet-wide 30-day estimate of exploitation activity, not a forecast for my systems). If they give no fixed versions, use the CVE record https://cveawg.mitre.org/api/cve/CVE-2026-43627 or the vendor advisory, not third-party mirrors. Tell me how to check whether I am affected and how to verify a fix. Cite the sources and say where they are uncertain.Opens a third-party assistant; its answer is its own, not cve.bz's. The prompt holds only this CVE's id and public links.
From the CNA's record in the CVE List.
| Product | Affected |
|---|---|
| ggml-org llama.cpp | from b1283 through b9058 |
| Distribution | Release | Package | Fixed in | Advisory |
|---|---|---|---|---|
| Debian | 14 | llama.cpp | no fix listed | DEBIAN-CVE-2026-43627 |
| Ubuntu | 26.04:LTS, 26.10 | llama.cpp | no fix listed | UBUNTU-CVE-2026-43627 |
From the distributions' data via OSV.dev: the Debian Security Tracker; Ubuntu Security Notices © Canonical Ltd., CC BY-SA 4.0. Modified: reformatted and linked to CVE records. Credits. On the distributions' own pages: Debian · Ubuntu.
Red Hat rates this Important.
Exploitation requires convincing a user or local process to execute inference workloads using maliciously oversized batch sizing parameters, leading to an arithmetic overflow during memory allocation and subsequent heap corruption. Full impact across Confidentiality, Integrity, and Availability (C:H, I:H, A:H) is possible if local memory execution controls fail, potentially resulting in arbitrary code execution within the application context. Deployments operating strictly with validated, bounded batch size inputs or running non-interactive server pipelines where batch parameters are hard-coded or strictly validated are unaffected.
To reduce the risk, avoid processing untrusted or malicious input with applications that use the llama.cpp library. Deploying applications that utilize llama.cpp within a sandboxed environment can further limit the potential impact of successful exploitation.
| Product | Package | State | Advisory |
|---|---|---|---|
| Red Hat Enterprise Linux AI (RHEL AI) 3 | rhelai3/bootc-cuda-rhel9 | Affected | |
| Red Hat Enterprise Linux AI (RHEL AI) 3 | rhelai3/bootc-gaudi-rhel9 | Will not fix | |
| Red Hat Enterprise Linux AI (RHEL AI) 3 | rhelai3/bootc-rocm-rhel9 | Affected | |
| Red Hat Enterprise Linux AI (RHEL AI) 3 | rhelai3/disk-image-cuda-rhel9 | Affected |
Contains Red Hat Security Data © Red Hat, Inc., licensed under CC BY 4.0. Source: https://access.redhat.com/security/data. Modified: reformatted. Red Hat's page for CVE-2026-43627.
The attack patterns MITRE's CAPEC relates to this record's weaknesses, as MITRE names them. Credits. More on the page of CWE-680 (11).