Medium [CVE-2026-73557] Incomplete remediation allows bypass via concurrent prompt processing
This medium-severity Red Hat Linux advisory covers CVE-2026-73557.
Android app · Google Play
Monitor future Red Hat Linux CVEs from your phone.
Choose a whole vendor or a precise platform, then receive matching security advisories by phone notification, email, or both. Coverage follows 32 official vendor sources and 160+ reviewed platform categories.
Summary
vLLM is an inference and serving engine for large language models.
From 0.20.2rc0 until 0.26.0, safe_load_prompt_embeds in vllm/renderers/embed_utils.py uses torch.sparse.check_sparse_tensor_invariants, whose process-global save, enable, and restore state can be raced by concurrent prompt_embeds parts submitted to POST /v1/chat/completions through AsyncMultiModalItemTracker.resolve_items, asyncio.gather, and the default executor, allowing an invalid sparse tensor to reach tensor.to_dense despite the CVE-2025-62164 guard when enable_prompt_embeds is enabled.
This issue is fixed in version 0.26.0. An incomplete remediation for a previous vulnerability (CVE-2025-62164) can be bypassed.
This occurs when concurrent prompt parts are processed, leading to a race condition in the handling of sparse tensors. This vulnerability could allow an attacker to cause a denial of service.
Red Hat severity: Moderate — CVSS 5.3 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L). Weakness: CWE-367.
- 0.20
Official advisory · high-confidence parse· fetched 2 hours ago·verify at source
Mitigation
Upgrade to a fixed release: 0.26.0. That is the remediation for this advisory.
The vendor advisory may list additional interim mitigations or workarounds not captured here — review it before change work.
Official advisory · high-confidence parse· fetched 2 hours ago·verify at source
Discussion(0)
No comments yet. Share field notes, upgrade gotchas, or questions — verify against the vendor advisory before acting on community advice.
Sign in to join the discussion.