CVE-2026-0599: Hugging Face Text Generation Inference Vulnerable to Uncontrolled Resource Consumption
Summary
CVE-2026-0599: Hugging Face Text Generation Inference vulnerable to Uncontrolled Resource Consumption. Affects text-generation versions < 3.3.7. Fixed in 3.3.7. How to fix ↓
Check if this vulnerability affects your application
Upload a manifest or SBOM and we'll show whether you're affected, plus every other risky package in the same scan.
No signup required · Results in about 60 seconds
Background
Hugging Face's Text Generation Inference is a powerful tool designed for generating text based on input prompts. It leverages advanced machine learning models to provide users with high-quality text generation capabilities. However, version 3.3.6 of this package is susceptible to a serious vulnerability that allows unauthenticated remote attackers to exploit unbounded external image fetching during input validation in VLM mode, leading to potential resource exhaustion.
Vulnerability Detail
The vulnerability arises when the router scans inputs for Markdown image links and performs a blocking HTTP GET request. This process reads the entire response body into memory and clones it before decoding, which can lead to resource exhaustion, including network bandwidth saturation, memory inflation, and CPU overutilization. Even if the request is later rejected for exceeding token limits, the vulnerability can still be triggered. The default deployment configuration, which lacks memory usage limits and authentication, exacerbates the impact, potentially crashing the host machine.
How To Fix
Minimum fixed version
3.3.7
This is the first release that contains the fix.
pip install --upgrade "text-generation==3.3.7"Tip — Not sure which version your app runs? Vulert can match this fix to it. Scan free →
Workaround
As a temporary workaround, consider implementing input validation to restrict the types of inputs that can be processed, or configure your deployment to include memory usage limits and authentication to mitigate the risk of resource exhaustion.
Conclusion
In summary, CVE-2026-0599 is a significant vulnerability in Hugging Face's Text Generation Inference that can lead to serious resource exhaustion issues. It is crucial to update to the patched version 3.3.7 to mitigate this risk. Additionally, using tools like Vulert can help monitor and alert for such vulnerabilities in real-time.
See every vulnerability recorded for text-generation →
FAQ
CVE-2026-0599 is a vulnerability in Hugging Face's Text Generation Inference that allows unauthenticated remote attackers to exploit unbounded external image fetching, leading to resource exhaustion.
You can check if your application is affected by using Vulert's playground at Vulert Scanner, which monitors and alerts for open-source vulnerabilities in real-time.
The potential consequences include network bandwidth saturation, memory inflation, CPU overutilization, and even crashing the host machine due to resource exhaustion.
Yes, the vulnerability is resolved in version 3.3.7 of the text-generation package. It is recommended to update to this version or later.
If you cannot update the package immediately, consider implementing input validation and configuring your deployment with memory usage limits and authentication as a temporary workaround.
Related vulnerabilities
Recent vulnerabilities recorded for text-generation.
