Read the docs
TI-2026-31732 Aliases CVE-2026-43628GHSA-wvf4-v63v-mwm5

llama.cpp b3978–b9058 Integer Underflow via DRY Sampler

Priority LOW 38CWE-125Published · Updated

Key facts

Attack vector
Local
Privileges required
None
User interaction
Passive
Attack complexity
Low
Attack requirements
None
Impact type
Code execution
Confidentiality · integrity · availability
High · High · High
Fix status
No fix listed

Attack profile from CVSS 4.0 · CNA. Fix status from CNA. Impact type from CWE-191 and the CVSS impact.

Summary

llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV
EPSS 0.00234 14th percentileScored 2026-11-07
Fix No fix listed No fixed version in the sources yet

SSVC decision points

CISA ADP · SSVC 2.0.3

Exploitation
None
Automatable
No
Technical impact
Total

The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.

Affected products

VendorProductAffected versionsFixed inStatusSource
ggml-orgllama.cppb3978 – b9058—No fix listedCNA
  • llama.cpp Vendorggml-org Affected b3978 – b9058 Fixed in — No fix listed CNA

CPE configurations NVD · 1

CPEVersionsRoleVendor
cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:*b3978 – b9058Vulnerableggml-org
  • cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:* Versions b3978 – b9058 Vulnerable ·ggml-org

Remediation

No fixed version listed by the sources yet.

Timeline

  1. 2026-10-02 Threatist record created from CVE-2026-43628
  2. 2026-10-03 Alias added: GHSA-wvf4-v63v-mwm5
  3. 2026-10-29 Affected versions changed · CNA

References

Patch 1

  • TI-2026-26459 llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend CVSS 3.1CRITICAL 9.8
  • TI-2024-28446 llama.cpp allows Arbitrary Address Read in rpc_server::get_tensor CVSS 3.1CRITICAL 9.8
  • TI-2026-17570 llama.cpp has Out-of-bounds Write in llama-server CVSS 3.1CRITICAL 9.8
  • TI-2024-28447 llama.cpp allows write-what-where in rpc_server::set_tensor CVSS 3.1CRITICAL 10.0
  • TI-2024-15627 Integer overflow in ggml-org llama.cpp CVSS 3.1CRITICAL 9.8
  • TI-2024-14482 Heap-based buffer overflow in ggml-org llama.cpp CVSS 3.1CRITICAL 9.8
See all 28 →

Sources

This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences