vLLM through 0.29.0 contains a denial of service vulnerability in the NIXL connector's prefix caching implementation that fails to properly validate block counts across multi-prompt completion requests in prefill/decode disaggregated deployments. Attackers can
| Vendor | Product | Version Range | Status |
|---|---|---|---|
| vllm-project | vllm | ≤ 0.29.0 |
affected |
Shenlong is analyzing...
Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.
| Vendor | Product | Affected Versions | CPE | Subscribe |
|---|---|---|---|---|
| vllm-project | vllm | 0 ~ 0.29.0 | - |
|
| # | POC Description | Source Link | Shenlong Link |
|---|
No public POC found.
Login to generate AI POC| CVE-2026-94627 | 7.5 HIGH | vLLM through 0.29.0 GPU KV Cache Leak via Mooncake Transfer ID Collision |
| CVE-2026-94624 | 7.5 HIGH | vLLM through 0.29.0 Denial of Service via Unbounded P2P KV Offloading Sessions |
| CVE-2026-94626 | 7.5 HIGH | vLLM through 0.29.0 Memory Exhaustion via Unvalidated NIXL tp_size |
| CVE-2026-94622 | 7.5 HIGH | vLLM through 0.29.0 Denial of Service via Incomplete NIXL KV Transfer Metadata |
| CVE-2026-94625 | 5.3 MEDIUM | vLLM through 0.29.0 Resource Exhaustion via Ownerless Mooncake Transfer Placeholders |
No comments yet