CVE-2026-78684: Uncontrolled Resource Consumption in vllm-project vllm
Description
vLLM versions before 0.27.0 have a vulnerability where DeepStream is not properly classified as a GPU backend, leading to omission of pixel-limit enforcement in the decode path. This allows unauthenticated attackers to activate DeepStream at request time, initializing the GPU decode pool and submitting video that bypasses resource controls, causing partial denial of service for concurrent requests.
CVSS v4.0
Score 6.9medium
Affected software
vllm-project
vllm
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The vulnerability in vLLM prior to version 0.27.0 involves improper classification of DeepStream as a GPU backend, which results in the failure to enforce pixel limits during video decoding. An unauthenticated attacker can exploit this by activating DeepStream at request time, which initializes a process-wide GPU decode pool. This allows the attacker to submit video data that bypasses resource controls, leading to uncontrolled resource consumption and causing partial denial of service for other concurrent requests.
Potential Impact
The impact is a partial denial of service condition affecting concurrent requests due to uncontrolled GPU resource consumption. This can degrade service availability and performance but does not require authentication or user interaction.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. No official fix or temporary mitigation is currently documented. Until a patch is available, monitoring for unusual GPU resource usage and restricting access to the affected service may help reduce risk.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-08-25T01:17:12.263Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6a8d820aacd9273b492253b4
Added to database: 08/25/2026, 11:52:42 UTC
Last enriched: 09/07/2026, 15:34:26 UTC
Last updated: 10/09/2026, 18:48:23 UTC
Views: 26
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.