CVE-2026-43632
NVD analysis in progress
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.
What this means
- Exposure
- Exploitable remotely over the network, without authentication and with no action from the victim. The attack does require particular conditions, which makes it less systematic.
- Impact
- An attacker can read sensitive data, modify or destroy data and take the service offline.
- Weakness
- The checked state changes before it is used, letting the attacker swap the target in between.
- Likelihood
- Its EPSS score stays low: nothing points to imminent exploitation, which is no reason to leave it unpatched.
What to doFold into the next patch cycle. Start with the instances exposed to the internet.
Read automatically from the CVSS vector, the weakness type (CWE) and the EPSS score. The technical description above remains the one published by NIST.