AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

GHSA-pr7f-p5mw-fc87: vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts

OSV AI package advisories · observation · Sep 4, 2026 · UTC

## Executive Summary The follow-up protection for CVE-2025-62164 is incomplete at vLLM revision `26587f9519e22a5c4549ead7595ad9ca3229c4fd`. It wraps serialized prompt-embedding reconstruction and dense conversion in `torch.sparse.check_sparse_tensor_invariants()`, but PyTorch 2.11.0 implements that context with save/enable/restore operations over process-global state. Two prompt-embedding parts in one `/v1/chat/completions` request are gathered concurrently on the event loop's default executor. When one context exits before the other loads its tensor, it can restore the global flag to `False`

Read original source ↗ Open in workspace

recordType
vulnerability
status
active
evidenceStatus
reported
region
Global

Evidence & attribution

First collected: 2026-09-20T22:31:48.298Z. This is not the publication date.