CVE-2026-34755
MediumCVSS 6.5Exploitation Probability (EPSS)
Low risk30th percentile - higher than 30% of all known CVEs
Summary
A vulnerability in vLLM from version 0.7.0 to before 0.19.0 allows an attacker to send a single API request with thousands of base64-encoded JPEG frames, causing the server to decode all frames into memory and crash with OOM. The issue is due to missing frame count limit in the VideoMediaIO.load_base64() method.
Risk Assessment
An attacker can remotely crash the vLLM server, leading to service disruption and potential denial of service for inference users.
Recommendation
Upgrade vLLM to version 0.19.0 or later immediately, which includes a fix that enforces a frame count limit in the base64 path.
Original NVD description (English source)
vLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base64() method at vllm/multimodal/media/video.py splits video/jpeg data URLs by comma to extract individual JPEG frames, but does not enforce a frame count limit. The num_frames parameter (default: 32), which is enforced by the load_bytes() code path, is completely bypassed in the video/jpeg base64 path. An attacker can send a single API request containing thousands of comma-separated base64-encoded JPEG frames, causing the server to decode all frames into memory and crash with OOM. This vulnerability is fixed in 0.19.0.