CVE-2026-43631: Use After Free in ggml-org llama.cpp
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
AI Analysis
Technical Summary
The vulnerability in llama.cpp (ggml-org) affects builds starting at b7492 and involves a use-after-free condition in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled. Attackers can remotely trigger this by sending requests while the server transitions to sleep mode, leading to concurrent worker threads accessing a freed vocab pointer. This pointer can be reclaimed with attacker-controlled data, enabling remote code execution. The CVSS 4.0 score is 9.2, indicating critical severity. There is no vendor advisory or patch information currently available, and no known exploits in the wild have been reported.
Potential Impact
Successful exploitation allows unauthenticated remote attackers to execute arbitrary code on the affected server, potentially leading to full system compromise. The vulnerability arises from a race condition causing use-after-free of a critical pointer during server sleep transitions.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, disabling the --sleep-idle-seconds feature or avoiding sending requests during server sleep transitions may reduce exposure.
CVE-2026-43631: Use After Free in ggml-org llama.cpp
Description
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
CVSS v4.0
Score 9.2critical
Affected software
ggml-org
llama.cpp
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The vulnerability in llama.cpp (ggml-org) affects builds starting at b7492 and involves a use-after-free condition in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled. Attackers can remotely trigger this by sending requests while the server transitions to sleep mode, leading to concurrent worker threads accessing a freed vocab pointer. This pointer can be reclaimed with attacker-controlled data, enabling remote code execution. The CVSS 4.0 score is 9.2, indicating critical severity. There is no vendor advisory or patch information currently available, and no known exploits in the wild have been reported.
Potential Impact
Successful exploitation allows unauthenticated remote attackers to execute arbitrary code on the affected server, potentially leading to full system compromise. The vulnerability arises from a race condition causing use-after-free of a critical pointer during server sleep transitions.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, disabling the --sleep-idle-seconds feature or avoiding sending requests during server sleep transitions may reduce exposure.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-05-01T18:22:45.641Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6a750702bf8831d5395f59ca
Added to database: 08/06/2026, 22:13:22 UTC
Last enriched: 08/14/2026, 15:54:49 UTC
Last updated: 09/21/2026, 22:01:34 UTC
Views: 29
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.