VAITP Dataset

← Back to the dataset

CVE-2026-112083

Weight Exporter DoS via crafted group_name in SGLang ≤0.5.19.

  • CVSS 5.5
  • CWE-404
  • Input Validation and Sanitization
  • Remote

A security flaw has been discovered in sgl-project SGLang up to 0.5.19. Affected is the function WeightExporter.send_weights_to_remote_instance of the file python/sglang/srt/model_executor/model_runner_components/weight_exporter.py of the component Weight Exporter. Performing a manipulation of the argument group_name results in denial of service. The attack can be initiated remotely. The exploit has been released to the public and may be used for attacks. The pull request to fix this issue awaits acceptance.

CVSS base score
5.5
Published
2026-10-19
OWASP
A08 Software and Data Integrity Failures
Orthogonal defect classification
Checking
Code defect classification
Missing Check
Category
Input Validation and Sanitization
Subcategory
Insecure Parsing or Deserialization
Accessibility scope
Remote
Impact
Denial of Service (DoS)
Affected component
Python
Fixed by upgrading
Yes

Solution

Upgrade sgl-project SGLang to version 0.5.20 or later, where the WeightExporter.send_weights_to_remote_instance function has been hardened against malicious group_name values.

Vulnerable code sample

import os

class WeightExporter:
    def __init__(self, base_dir: str):
        self.base_dir = base_dir

    def send_weights_to_remote_instance(self, group_name: str, weights: bytes):
        # VULNERABLE: unsanitized group_name used in path construction
        path = os.path.join(self.base_dir, f"{group_name}.weights")
        with open(path, "wb") as f:
            f.write(weights)
        # Simulate remote send (placeholder)
        self._remote_send(path)

    def _remote_send(self, file_path: str):
        # Placeholder for remote transmission logic
        pass

Patched code sample

import os

class WeightExporter:
    def __init__(self, base_dir: str):
        self.base_dir = base_dir
        self._allowed_groups = {"default", "experiment", "test"}

    def send_weights_to_remote_instance(self, group_name: str, weights: bytes):
        # FIX: validate group_name against whitelist to prevent path abuse
        if group_name not in self._allowed_groups:
            raise ValueError("Invalid group_name")
        path = os.path.join(self.base_dir, f"{group_name}.weights")
        with open(path, "wb") as f:
            f.write(weights)
        self._remote_send(path)

    def _remote_send(self, file_path: str):
        # Placeholder for remote transmission logic
        pass

Published fix

Public sources link this CVE to the changes below. The code samples above were generated to illustrate the defect; these links lead to the code the project actually changed.

Cite this entry

@misc{vaitp:cve2026112083,
  title        = {{Weight Exporter DoS via crafted group_name in SGLang ≤0.5.19.}},
  author       = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
  year         = {2026},
  note         = {VAITP Python Vulnerability Dataset, entry CVE-2026-112083},
  howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2026-112083/}}
}

Download this entry as JSON Download the whole dataset

Introducing the "VAITP dataset": a specialized repository of Python vulnerabilities and patches, meticulously compiled for the use of the security research community. As Python's prominence grows, understanding and addressing potential security vulnerabilities become crucial. Crafted by and for the cybersecurity community, this dataset offers a valuable resource for researchers, analysts, and developers to analyze and mitigate the security risks associated with Python. Through the comprehensive exploration of vulnerabilities and corresponding patches, the VAITP dataset fosters a safer and more resilient Python ecosystem, encouraging collaborative advancements in programming security.

The supreme art of war is to subdue the enemy without fighting.

Sun Tzu – “The Art of War”

:: Shaping the future through research and ingenuity ::

Before you use the VAITP AI models

Please read and accept the following disclaimer:

By choosing "I understand and accept", you acknowledge that you understand and agree to these terms.