CVE-2024-34359 is a critical-severity (CVSS 9.6) CWE-76 vulnerability in llama-cpp-python. EPSS puts its 30-day exploitation probability at 26.0% (98th percentile). A fix is available for llama-cpp-python — see the affected versions and patch details below.
llama-cpp-python vulnerable to Remote Code Execution by Server-Side Template Injection in Model Metadata
Exploitation Status
Proof-of-concept exploit code exists
- CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
- A successful exploit gives an attacker total control of the affected component, not partial access.
Exploitation and automatability from CISA’s SSVC triage for CVE-2024-34359.
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
How urgent is this, really
CVE-2024-34359 plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.
Where this sits among everything scored
Of 376,715 CVEs with a current EPSS score, this one falls in the 10–50% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.
Real-World Exposure
llama-cpp-pythonReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.
Description
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 Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐍PyPI | llama-cpp-python | ≥ 0.2.30&&< 0.2.72 | 0.2.72pip install --upgrade 'llama-cpp-python==0.2.72' |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for llama-cpp-python, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update llama-cpp-python to 0.2.72 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2024-34359 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like CVE-2024-34359 can be triaged on real exposure rather than presence alone.
Tailored to CVE-2024-34359. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is CVE-2024-34359 in your dependencies?
O3 Security finds CVE-2024-34359 across PyPI dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.