CVE-2026-73557
Published:August 13, 2026
Updated:August 31, 2026
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.
Affected Packages
vllm (CONDA):
Affected version(s) >=0.21.0 <0.26.0Fix Suggestion:
Update to version 0.26.0https://github.com/vllm-project/vllm.git (GITHUB):
Affected version(s) >=v0.20.2 <v0.26.0Fix Suggestion:
Update to version v0.26.0vllm (PYTHON):
Affected version(s) >=0.20.2 <0.26.0Fix Suggestion:
Update to version 0.26.0Related Resources (6)
Do you need more information?
Contact UsCVSS v4
Base Score:
6.3
Attack Vector
NETWORK
Attack Complexity
LOW
Attack Requirements
PRESENT
Privileges Required
NONE
User Interaction
NONE
Vulnerable System Confidentiality
NONE
Vulnerable System Integrity
NONE
Vulnerable System Availability
LOW
Subsequent System Confidentiality
NONE
Subsequent System Integrity
NONE
Subsequent System Availability
NONE
CVSS v3
Base Score:
5.3
Attack Vector
NETWORK
Attack Complexity
LOW
Privileges Required
NONE
User Interaction
NONE
Scope
UNCHANGED
Confidentiality
NONE
Integrity
NONE
Availability
LOW
Weakness Type (CWE)
Concurrent Execution using Shared Resource with Improper Synchronization ('Race Condition')
EPSS
Base Score:
0.25