CVE-2024-34359

GHSA-56xg-wfcc-g829 · published 10 May 2024 · updated 16 Nov 2026

llama-cpp-python vulnerable to Remote Code Execution by Server-Side Template Injection in Model Metadata

10.0CriticalCVSS 9.7EPSS 26%

No exploitation in the wild has been reported. EPSS puts the chance of exploitation in the next 30 days at 26%. Fixed in llama-cpp-python 0.2.72.

llama-cpp-python is the Python bindings for llama.cpp. `llama-cpp-python` depends on class `Llama` in `llama.py` to load `.gguf` llama.cpp or Latency Machine Learning Models. The `__init__` constructor built in the `Llama` takes several parameters to configure the loading and running of the model. Other than `NUMA, LoRa settings`, `loading tokenizers,` and `hardware settings`, `__init__` also loads the `chat template` from targeted `.gguf` 's Metadata and furtherly parses it to `llama_chat_format.Jinja2ChatFormatter.to_chat_handler()` to construct the `self.chat_handler` for this model. Nevertheless, `Jinja2ChatFormatter` parse the `chat template` within the Metadate with sandbox-less `jinja2.Environment`, which is furthermore rendered in `__call__` to construct the `prompt` of interaction. This allows `jinja2` Server Side Template Injection which leads to remote code execution by a carefully constructed payload.

Affected

Vendors, products and packages with version ranges

VendorProductVersionsFixed
abetlen>=0.2.30 <=0.2.71
pypi: llama-cpp-python>=0.2.30 <0.2.72, >=0.2.30 <=0.2.710.2.72

Timeline

Every event on this vulnerability

Embed

A live severity badge for an advisory, README or ticket

CVE-2024-34359 severity on Zeptric[![CVE-2024-34359](https://zeptric.io/badge/cve/CVE-2024-34359.svg)](https://zeptric.io/cve/CVE-2024-34359)