CVE-2026-43630: CWE-125 Out-of-bounds Read in ggml-org llama.cpp
An out-of-bounds read vulnerability exists in llama.cpp builds b5702 through b7653 in the recurrent memory state restore path. This flaw allows attackers with write access to the slot save directory to craft malicious slot files that trigger reading beyond allocated memory, leaking heap data including pointer values. The vulnerability can expose sensitive memory contents in server logs, undermining ASLR protections and potentially aiding further exploitation. The CVSS 4.0 base score is 6.3, indicating medium severity. No official patch or remediation guidance is currently available.
AI Analysis
Technical Summary
llama.cpp versions b5702 and 0.17.1 contain an out-of-bounds read vulnerability (CWE-125) in the recurrent memory state restore path. Attackers with write permissions to the slot save directory can create a malicious slot file with an oversized seq_id value, causing the software to read memory beyond the allocated cells array. This results in leakage of heap data, including pointer values, into server logs. Such leakage defeats Address Space Layout Randomization (ASLR) protections and may facilitate further exploitation. The vulnerability has a CVSS 4.0 score of 6.3 (medium severity). There is no vendor advisory or patch available at this time.
Potential Impact
The vulnerability allows an attacker with write access to a specific directory to cause an out-of-bounds read, leaking heap memory contents including pointer values. This memory disclosure can defeat ASLR protections, potentially making subsequent attacks easier. However, exploitation requires write access to the slot save directory, limiting the attack surface. There are no known exploits in the wild. The impact is classified as medium severity based on the CVSS score.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, restrict write access to the slot save directory to trusted users only to prevent attackers from placing malicious slot files.
CVE-2026-43630: CWE-125 Out-of-bounds Read in ggml-org llama.cpp
Description
An out-of-bounds read vulnerability exists in llama.cpp builds b5702 through b7653 in the recurrent memory state restore path. This flaw allows attackers with write access to the slot save directory to craft malicious slot files that trigger reading beyond allocated memory, leaking heap data including pointer values. The vulnerability can expose sensitive memory contents in server logs, undermining ASLR protections and potentially aiding further exploitation. The CVSS 4.0 base score is 6.3, indicating medium severity. No official patch or remediation guidance is currently available.
CVSS v4.0
Score 6.3medium
Affected software
pkg:github/ggml-org/llama.cppRun on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Weaknesses
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
llama.cpp versions b5702 and 0.17.1 contain an out-of-bounds read vulnerability (CWE-125) in the recurrent memory state restore path. Attackers with write permissions to the slot save directory can create a malicious slot file with an oversized seq_id value, causing the software to read memory beyond the allocated cells array. This results in leakage of heap data, including pointer values, into server logs. Such leakage defeats Address Space Layout Randomization (ASLR) protections and may facilitate further exploitation. The vulnerability has a CVSS 4.0 score of 6.3 (medium severity). There is no vendor advisory or patch available at this time.
Potential Impact
The vulnerability allows an attacker with write access to a specific directory to cause an out-of-bounds read, leaking heap memory contents including pointer values. This memory disclosure can defeat ASLR protections, potentially making subsequent attacks easier. However, exploitation requires write access to the slot save directory, limiting the attack surface. There are no known exploits in the wild. The impact is classified as medium severity based on the CVSS score.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, restrict write access to the slot save directory to trusted users only to prevent attackers from placing malicious slot files.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-05-01T18:22:45.641Z
- Cvss Version
- 4.0
- State
- PUBLISHED
- Remediation Level
- null
Threat ID: 6a750702bf8831d5395f59c5
Added to database: 08/06/2026, 22:13:22 UTC
Last enriched: 08/06/2026, 23:12:48 UTC
Last updated: 08/07/2026, 00:01:00 UTC
Views: 3
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.