Skip to content

CVE-2026-0599 Unbounded External Image Fetch in Validation Leads to Resource-Exhaustion DoS in huggingface/text-generation-inference

HighCVSS 7.5 EPSS 29.5%
Published
Updated
Exploit status
No public exploit known

Description

A vulnerability in huggingface/text-generation-inference version 3.3.6 allows unauthenticated remote attackers to exploit unbounded external image fetching during input validation in VLM mode. The issue arises when the router scans inputs for Markdown image links and performs a blocking HTTP GET request, reading the entire response body into memory and cloning it before decoding. This behavior can lead to resource exhaustion, including network bandwidth saturation, memory inflation, and CPU overutilization. The vulnerability is triggered even if the request is later rejected for exceeding token limits. The default deployment configuration, which lacks memory usage limits and authentication, exacerbates the impact, potentially crashing the host machine. The issue is resolved in version 3.3.7.

Affected products

ProductAffected versionsFixed in
pypi text-generation(PyPI) < 3.3.7 3.3.7
pypi text-generation(PyPI) = 0.1.0 No fix listed
pypi text-generation(PyPI) = 0.2.0 No fix listed
pypi text-generation(PyPI) = 0.2.1 No fix listed
pypi text-generation(PyPI) = 0.3.0 No fix listed
pypi text-generation(PyPI) = 0.3.1 No fix listed
pypi text-generation(PyPI) = 0.4.0 No fix listed
pypi text-generation(PyPI) = 0.4.1 No fix listed
pypi text-generation(PyPI) = 0.5.0 No fix listed
pypi text-generation(PyPI) = 0.5.1 No fix listed
pypi text-generation(PyPI) = 0.5.2 No fix listed
pypi text-generation(PyPI) = 0.6.0 No fix listed
pypi text-generation(PyPI) = 0.6.1 No fix listed
pypi text-generation(PyPI) = 0.7.0 No fix listed
huggingface/text-generation-inference < 3.3.7 3.3.7

References