CyberRota Analysis
AI-GeneratedThe vulnerability affects the vLLM inference and serving engine for large language models, specifically versions 0.20.2rc0 to 0.26.0, where a race condition in the handling of sparse tensors can lead to invalid data being processed. This could result in unexpected behavior or crashes when processing prompt embeddings in concurrent environments. Organizations utilizing vLLM for large language model deployments should prioritize upgrading to version 0.26.0 to mitigate potential risks associated with this vulnerability.
Public Exploit Signal
A public exploit, PoC, GitHub repository or Metasploit reference was detected for this CVE.
Note: these links are listed for security research and verification purposes only.
Original NVD Description
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.