CVE-2026-100654: vLLM before 0.29.0 Denial of Service via out-of-range stop_token_ids

Vulnerability summary

CVE-2026-100654 is a high (CVSS 7.1) Improper Validation of Array Index vulnerability in vllm-project vllm. EPSS estimates a 0.31% chance of exploitation in the next 30 days, higher than 23% of scored vulnerabilities.

Timeline

1. 26-09-2026
   PublishedVulnerability first disclosed.
2. 30-09-2026
   Last modifiedVulnerability information updated.

Description

vLLM before 0.29.0 accepts user-controlled stop_token_ids on the OpenAI-compatible POST /v1/completions and POST /v1/chat/completions endpoints but validates only that the values are integers, not that each token id is within the model vocabulary/logits range. When min_tokens > 0, the stop token ids are used as logits indices to suppress stop tokens, so an out-of-range id reaches a CUDA indexing operation (index_put_) and triggers a device-side assertion. An authenticated API user can send a single malformed completion request that returns 500 Internal Server Error and puts EngineCore into a fatal state, causing subsequent requests to fail until the service is restarted (denial of service).

• CWE-129Improper Validation of Array IndexThe product uses untrusted input when calculating or using an array index, but the product does not validate or incorrectly validates the index to ensure the index references a valid position within the array.