Skip to main content
CVE-2026-70640VulnerabilityPublished Updated

llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context

Source GCVE Database

0
Compare
Authoritative record:CVE-2026-70640· High 7.3

Not known to be exploited · CVSS 7.3 High.

Data as of 14 Nov 2026, 01:04 UTC

CVSS 4.0 base score
7.3High
How severe the flaw is on the 0–10 CVSS scale.

Thin evidence

One scorer or none
CVSS scores by source
CNA (VulnCheck)CVSS 4.07.3
CNA (VulnCheck)CVSS 3.17.0
CVSS:4.0/AV:L/AC:H/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 6 Aug 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponized: no evidence on record.

    No public exploit code indexed in Metasploit or Exploit-DB

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 13 Nov 2026.

What to do

Affected software

ProductAffectedFixed in
ggml-orgllama.cppAffected
  • >=b1886 <=b7445
Fixed in–
GitHubggml-org/llama.cpppkg:github/ggml-org/llama.cppOther threatsAffected
  • =0.9.0
Fixed in–

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

Description

llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.

References

Community reviews

Loading reviews…