critical CVSS 9.8

CVE-2025-23316·Python vulnerability

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend, where an attacker could cause a remote code execution by manipulating the model name parameter in the model control APIs. A successful exploit of this vulnerability might lead to remote code execution, denial of service, information disclosure, and data tampering.

Severity
critical
Software
Python
Fixed in
25.08
Published
2025-09-17

Affected versions

Until: 25.08

Fixed in: 25.08

How to fix this CVE

Update NVIDIA Triton Inference Server to version 25.08 or later to patch the Python backend model name parameter vulnerability. Organizations running Triton with the Python backend should prioritize this update immediately, as the vulnerability allows unauthenticated remote code execution. Verify the updated version is deployed across all inference servers before resuming production traffic.

sudo dnf check-update python3 && sudo dnf upgrade python3

Defensia detects this vulnerability

What an exploitation attempt looks like

Sample log line indicative of exploitation attempts:

POST requests to /v2/models/{MODEL_NAME}/config or /v2/models/{MODEL_NAME}/load with model name parameters containing shell metacharacters (;|&$`\n) or path traversal sequences (../) should be flagged as potential exploitation attempts

WAF mitigation (if patching is not yet possible)

Add this rule to your WAF to block exploitation attempts while you schedule the patch. 

Implement strict input validation on all model name parameters accepted by Triton's model control APIs, restricting to alphanumeric characters, hyphens, and underscores only. Rate-limit requests to /v2/models/*/config and /v2/models/*/load endpoints to 10 requests per minute per source IP.

How to check if you are affected

  1. Run 'python3 --version' to confirm the current Python version installed on the system
  2. Check if NVIDIA Triton Inference Server is running via 'ps aux | grep tritonserver' and verify the deployment version
  3. Review Triton server logs for unexpected model loading attempts or malformed model names: 'grep -i "model" /var/log/triton/*.log | grep -E "(error|exception|traceback)"'
  4. Confirm the Triton version is 25.08 or later by checking the server info endpoint: 'curl -s http://localhost:8000/v2/health/ready | grep -o version'

FAQ

What is CVE-2025-23316?

CVE-2025-23316 is a critical remote code execution vulnerability in NVIDIA Triton Inference Server's Python backend that allows attackers to inject malicious code through crafted model name parameters in the model control APIs, bypassing authentication entirely.

Is CVE-2025-23316 being actively exploited?

According to CISA, this vulnerability is not currently listed in the Known Exploited Vulnerabilities (KEV) catalog and no public exploits are available, though the critical CVSS score of 9.8 warrants immediate patching.

What versions of Python are affected by CVE-2025-23316?

All versions of NVIDIA Triton Inference Server up to and including version 25.07 that use the Python backend are affected; version 25.08 and later contain the fix.

How do I check if my server is vulnerable to CVE-2025-23316?

Run 'tritonserver --version' to confirm your Triton version; if it is below 25.08, your system is vulnerable. Additionally, check if the Python backend is enabled in your model configuration files.

Does Defensia detect CVE-2025-23316?

Yes — Defensia's CVE advisory scanner compares installed package versions against the NVD database. If Python or NVIDIA Triton is installed on a monitored server, CVE-2025-23316 will appear in your dashboard with remediation steps.

Related Python CVEs

CVE-2026-34938CVSS 10PraisonAI is a multi-agent teams system. Prior to version 1.5.90, execute_code() in praisonai-agents runs attacker-controlled Python inside a three-layer sandbox that can be fully bypassed by passing a str subclass with an overridden startswith() method to the _safe_getattr wrapper, achieving arbitrary OS command execution on the host. This issue has been patched in version 1.5.90.
CVE-2026-28505CVSS 10Tautulli is a Python based monitoring and tracking tool for Plex Media Server. Prior to version 2.17.0, the str_eval() function in notification_handler.py implements a sandboxed eval() for notification text templates. The sandbox attempts to restrict callable names by inspecting code.co_names of the compiled code object. However, co_names only contains names from the outer code object. When a lambda expression is used, it creates a nested code object whose attribute accesses are stored in code.co_consts, NOT in code.co_names. The sandbox never inspects nested code objects. This issue has been patched in version 2.17.0.
CVE-2026-32871CVSS 10FastMCP is a Pythonic way to build MCP servers and clients. Prior to version 3.2.0, the OpenAPIProvider in FastMCP exposes internal APIs to MCP clients by parsing OpenAPI specifications. The RequestDirector class is responsible for constructing HTTP requests to the backend service. A vulnerability exists in the _build_url() method. When an OpenAPI operation defines path parameters (e.g., /api/v1/users/{user_id}), the system directly substitutes parameter values into the URL template string without URL-encoding. Subsequently, urllib.parse.urljoin() resolves the final URL. Since urljoin() interprets ../ sequences as directory traversal, an attacker controlling a path parameter can perform path traversal attacks to escape the intended API prefix and access arbitrary backend endpoints. This results in authenticated SSRF, as requests are sent with the authorization headers configured in the MCP provider. This issue has been patched in version 3.2.0.
CVE-2026-33054CVSS 10Mesop is a Python-based UI framework that allows users to build web applications. Versions 1.2.2 and below contain a Path Traversal vulnerability that allows any user supplying an untrusted state_token through the UI stream payload to arbitrarily target files on the disk under the standard file-based runtime backend. This can result in application denial of service (via crash loops when reading non-msgpack target files as configurations), or arbitrary file manipulation. This vulnerability heavily exposes systems hosted utilizing FileStateSessionBackend. Unauthorized malicious actors could interact with arbitrary payloads overwriting or explicitly removing underlying service resources natively outside the application bounds. This issue has been fixed in version 1.2.3.
CVE-2026-25632CVSS 10EPyT-Flow is a Python package designed for the easy generation of hydraulic and water quality scenario data of water distribution networks. Prior to 0.16.1, EPyT-Flow’s REST API parses attacker-controlled JSON request bodies using a custom deserializer (my_load_from_json) that supports a type field. When type is present, the deserializer dynamically imports an attacker-specified module/class and instantiates it with attacker-supplied arguments. This allows invoking dangerous classes such as subprocess.Popen, which can lead to OS command execution during JSON parsing. This also affects the loading of JSON files. This vulnerability is fixed in 0.16.1.

References

Track CVEs across your fleet automatically

Defensia scans your Linux servers and tells you exactly which ones are running vulnerable versions — including CVE-2025-23316. Free for 1 server.

Get started free