CVE-2026-93436 vllm-project CVE debrief

Who should care

Defenders responsible for vLLM deployments should assess exposure and prioritize verification and updates to prevent memory exhaustion attacks. This includes operators, platform administrators, vulnerability management teams, and security teams. They should verify vLLM versions and deployments to assess exposure and implement compensating controls for exposed systems while remediation is scheduled and verified.

Why it matters

CVE-2026-93436 is a high-severity vulnerability in vLLM that can lead to memory exhaustion and denial-of-service conditions. Defenders should prioritize verification and updates to prevent attacks.

• Memory exhaustion can lead to denial-of-service conditions
• Decode-worker restarts may be required to recover from attacks
• Verification of vLLM versions and deployments is necessary
• Remediation priority is high due to potential for repeated attacks

Technical summary

vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with max_tokens=0 to exhaust decode-worker memory without bound until the worker restarts. This vulnerability affects vLLM deployments, particularly those with prefill/decode disaggregated deployments. Defenders should prioritize verifying and updating vLLM deployments to prevent memory exhaustion attacks. The vulnerability allows remote attackers to submit requests with max_tokens=0 to exhaust decode-worker memory without bound until the worker restarts.

Defensive priority

Defenders should prioritize verifying and updating vLLM deployments to prevent memory exhaustion attacks.

Recommended defensive actions

• Verify vLLM deployments for version 0.29.0 or earlier
• Update vLLM to a version beyond 0.29.0 if available
• Monitor decode-worker memory usage for unusual patterns
• Implement rate limiting on inference requests
• Review compensating controls for exposed systems while remediation is scheduled and verified
• Check relevant monitoring, detection, and logs for exposed assets that need extra review
• Track exceptions, retest remediated assets, and close the item only after evidence is documented

Evidence notes

The CVE record and NVD entry provide details on the vulnerability, but the scope of affected versions and deployments requires verification. Defenders should verify vLLM versions and deployments to assess exposure. The vulnerability allows remote attackers to submit requests with max_tokens=0 to exhaust decode-worker memory without bound until the worker restarts. Evidence of exploitation may include unusual memory usage patterns and decode-worker restarts.

Sources and references

Methodology and review provenance

AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.