llama.cpp b5899 before b10261 Heap Out-of-Bounds Write via plamo2 Tokenizer

Summary

llama.cpp b5899 before b10261 contains a heap out-of-bounds write vulnerability in llm_tokenizer_plamo2::build that allows attackers to corrupt heap memory via crafted GGUF models. Attackers can supply BYTE-typed token text like <0x-1> whose negative parsed value writes token ids before the bytes_ buffer during model loading, causing crashes or potential code execution.

Remediation

1. Inventory every ggml-org llama.cpp deployment, including staging.
2. Upgrade ggml-org llama.cpp to b10261.
3. Confirm the running version after the upgrade.

References

• Patchhttps://github.com/ggml-org/llama.cpp/commit/cf06ad7dfe79e7bf2934150d4fefe1ce8db05490
• Patchhttps://github.com/ggml-org/llama.cpp/pull/25969
• Patchhttps://github.com/ggml-org/llama.cpp/pull/26511
• Advisoryhttps://www.vulncheck.com/advisories/llama-cpp-b5899-before-b10261-heap-out-of-bounds-write-via-plamo2-tokenizer
• Referencehttps://github.com/ggml-org/llama.cpp/blob/b10259/src/llama-vocab.cpp#L1376-L1377
• Referencehttps://github.com/ggml-org/llama.cpp/releases/tag/b10261