SOURCE-LINKED INTELLIGENCE
GHSA-83vm-p52w-f9pw: vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters
### Summary The `extract_hidden_states` speculative decoding proposer in vLLM returns a tensor with an incorrect shape after the first decode step, causing a `RuntimeError` that crashes the EngineCore process. The crash is triggered when any request in the batch uses sampling penalty parameters (`repetition_penalty`, `frequency_penalty`, or `presence_penalty`). A single request with a penalty parameter (e.g., `"repetition_penalty": 1.1`) is sufficient to crash the server. The crash is deterministic and immediate — no concurrency, race condition, or special workload is required. ### Details In
Read original source ↗ Open in workspace
- recordType
- vulnerability
- status
- active
- evidenceStatus
- reported
- region
- Global
Evidence & attribution
- OSV AI package advisories · 2026-05-06T21:45:51.000Z
First collected: 2026-09-20T22:31:48.298Z. This is not the publication date.