{"id":"CVE-2026-43632","state":"PUBLISHED","title":"llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints","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.","publishedAt":"2026-08-06T15:53:19.379Z","updatedAt":"2026-08-14T16:49:56.563Z","severity":{"score":9.2,"label":"Critical","version":"4.0","vector":"CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","source":"VulnCheck"},"assigner":"VulnCheck","products":["ggml-org llama.cpp"],"affected":[{"vendor":"ggml-org","product":"llama.cpp","defaultStatus":"affected","versions":[{"version":"b7492","status":"affected","lessThanOrEqual":"b9060","versionType":"custom"}]}],"references":["https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches","https://www.vulncheck.com/advisories/llama-cpp-b7492-b9060-use-after-free-in-tokenization-endpoints"],"weaknesses":["CWE-416","CWE-367"],"ingestedAt":"2026-10-29T20:11:57.932Z","source":{"repository":"CVEProject/cvelistV5","path":"cves/2026/43xxx/CVE-2026-43632.json","commit":"556c9d3a8cd17e218bb8e452ba7cd619a384ff47"},"raw":{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-43632","state":"PUBLISHED","datePublished":"2026-08-06T15:53:19.379Z","dateUpdated":"2026-08-14T16:49:56.563Z","assignerShortName":"VulnCheck","assignerOrgId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","dateReserved":"2026-05-01T18:22:45.641Z"},"containers":{"cna":{"title":"llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints","descriptions":[{"lang":"en","value":"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."}],"affected":[{"vendor":"ggml-org","product":"llama.cpp","defaultStatus":"affected","versions":[{"version":"b7492","status":"affected","lessThanOrEqual":"b9060","versionType":"custom"}]}],"references":[{"url":"https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches","tags":["mitigation"]},{"url":"https://www.vulncheck.com/advisories/llama-cpp-b7492-b9060-use-after-free-in-tokenization-endpoints","tags":["third-party-advisory"]}],"metrics":[{"format":"CVSS","scenarios":[{"lang":"en","value":"GENERAL"}],"cvssV4_0":{"attackVector":"NETWORK","attackComplexity":"HIGH","attackRequirements":"PRESENT","privilegesRequired":"NONE","userInteraction":"NONE","vulnConfidentialityImpact":"HIGH","subConfidentialityImpact":"NONE","vulnIntegrityImpact":"HIGH","subIntegrityImpact":"NONE","vulnAvailabilityImpact":"HIGH","subAvailabilityImpact":"NONE","exploitMaturity":"NOT_DEFINED","Safety":"NOT_DEFINED","Automatable":"NOT_DEFINED","Recovery":"NOT_DEFINED","valueDensity":"NOT_DEFINED","vulnerabilityResponseEffort":"NOT_DEFINED","providerUrgency":"NOT_DEFINED","version":"4.0","baseSeverity":"CRITICAL","baseScore":9.2,"vectorString":"CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N"}},{"format":"CVSS","scenarios":[{"lang":"en","value":"GENERAL"}],"cvssV3_1":{"attackVector":"NETWORK","attackComplexity":"HIGH","privilegesRequired":"NONE","userInteraction":"NONE","scope":"UNCHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH","version":"3.1","baseSeverity":"HIGH","baseScore":8.1,"vectorString":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H"}}],"problemTypes":[{"descriptions":[{"cweId":"CWE-416","lang":"en","description":"Use After Free","type":"CWE"},{"cweId":"CWE-367","lang":"en","description":"Time-of-check Time-of-use (TOCTOU) Race Condition","type":"CWE"}]}],"providerMetadata":{"shortName":"VulnCheck","orgId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","dateUpdated":"2026-08-14T16:49:56.563Z"},"tags":["x_open-source"],"datePublic":"2026-05-06T00:00:00.000Z","cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:*","versionStartIncluding":"b7492","versionEndIncluding":"b9060"}]}]}],"credits":[{"lang":"en","value":"Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera","type":"finder"},{"lang":"en","value":"Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera","type":"finder"}],"source":{"discovery":"EXTERNAL"},"x_generator":{"engine":"vulncheck"}},"adp":[{"title":"CISA ADP Vulnrichment","metrics":[{"other":{"type":"ssvc","content":{"timestamp":"2026-08-08T02:09:08.150856Z","id":"CVE-2026-43632","options":[{"Exploitation":"none"},{"Automatable":"no"},{"Technical Impact":"total"}],"role":"CISA Coordinator","version":"2.0.3"}}}],"providerMetadata":{"shortName":"CISA-ADP","orgId":"134c704f-9b21-4f2e-91b3-4a467353bcc0","dateUpdated":"2026-08-08T02:09:18.676Z"}}]}}}