Read the docs
TI-2024-39152 Aliases CVE-2024-58400GHSA-7wfj-q343-9m2r

llama.cpp Out-of-Bounds Read in llama_decode via Token IDs

Priority MEDIUM 44CWE-129Published · Updated

Key facts

Attack vector
Network
Privileges required
None
User interaction
None
Attack complexity
Low
Attack requirements
None
Impact type
—
Confidentiality · integrity · availability
Low · None · High
Fix status
Fixed

Attack profile from CVSS 4.0 · CNA. Fix status from CNA.

Summary

llama.cpp from 0 before b3701 contains an out-of-bounds read in llama_decode_internal and llama_encode_internal. The code does not validate token IDs, so an improper array index check (CWE-129) lets reads go past the embedding memory. A remote unauthenticated attacker can send integer-array prompts to the llama-server /completion endpoint with IDs outside 0 to n_vocab to read memory or crash the process.

Written with Claude Haiku 5.5 from the sources below · 2026-10-28

Description from the CNA

llama.cpp before b3701 contains an out-of-bounds read vulnerability in llama_decode_internal and llama_encode_internal that allows remote unauthenticated attackers to read memory by supplying unvalidated token IDs. Attackers can send integer-array prompts to the llama-server /completion endpoint with IDs outside 0 to n_vocab to read embedding memory out of bounds or crash the process.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV
EPSS 0.00422 36th percentileScored 2026-11-07
Fix Fixed llama.cpp (ggml-org): b3701

SSVC decision points

CISA ADP · SSVC 2.0.3

Exploitation
None
Automatable
Yes
Technical impact
Partial

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.cppbefore b3701b3701Patch availableCNA
  • llama.cpp Vendorggml-org Affected before b3701 Fixed in b3701 Patch available CNA

CPE names Threatist (inferred) · 1

Inferred from how NVD named this product before; not given by a source.

CPEVersionsRoleVendor
cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:*——ggml-org
  • cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:* Versions — — ·ggml-org

Remediation

Update llama.cpp to b3701, the fixed version.

Written with Claude Haiku 5.5 from the sources below · 2026-10-28

  • Fix · CNA llama.cpp (ggml-org) Fixed in b3701

Timeline

  1. 2026-10-27 Threatist record created from CVE-2024-58400
  2. 2026-10-27 CISA SSVC decision points changed
  3. 2026-10-27 Alias added: GHSA-7wfj-q343-9m2r

References

Patch 3

Other 2

  • 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 52 →

Sources

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