CVE-2026-94626 PUBLISHED

vLLM through 0.29.0 fails to validate the tp_size parameter in kv_transfer_params on OpenAI-compatible completion endpoints, allowing attackers to allocate unbounded memory. Attackers can supply arbitrary tp_size values in prefill/decode disaggregated deployments to exhaust memory and trigger kernel OOM-kill of the decode worker process.

• Mingkai Yu finder
• Jiapeng Li finder
• Jiajia Liu finder

• Pull Request #51137
• https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/utils.py#L569-L573
• https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/v1/nixl/metadata.py#L277
• https://github.com/vllm-project/vllm
• VulnCheck Advisory: vLLM through 0.29.0 Memory Exhaustion via Unvalidated NIXL tp_size

• Memory Allocation with Excessive Size Value CWE