CVE-2026-43632
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task…
Does this matter?
High impact if exploited, but EPSS currently rates exploitation as unlikely (0.33%). Schedule it in the normal patch cycle and watch for a rise in EPSS or a public exploit.
Description
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.
- CVSS 4.0
- 9.2 CRITICALCVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/M
- EPSS
- 0.33% probability · 26th percentile
- CISA KEV
- Not listed
- Weakness
- CWE-367, CWE-416
- Source
- disclosure@vulncheck.com
References
Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.