Goal Reached Thanks to every supporter — we hit 100%!

Goal: 1000 CNY · Raised: 1359 CNY

100%
Buy Us a Coffee

CVE-2026-73557— vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts

Quick assessment

Affected
vllm-project vllm
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.

vLLM是vLLM团队开源的一个适用于 LLM 的高吞吐量和内存高效推理和服务引擎。 vLLM 0.20.2rc0版本至0.26.0之前版本存在竞争条件问题漏洞，该漏洞源于并发提交的prompt_embeds部分与torch.sparse.check_sparse_tensor_invariants的进程全局状态保存、启用和恢复操作产生竞争，可能导致无效稀疏张量到达tensor.to_dense，绕过CVE-2025-62164防护。

CVSS 6.3 · Medium EPSS 0.40% · P33

Possible ATT&CK Techniques 1 AI

T1190 · Exploit Public-Facing Application

Affected Version Matrix 1

VendorProduct Version RangeStatus
vllm-project vllm >= 0.20.2rc0, < 0.26.0 affected
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-73557

Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗

Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.

Vulnerability Title
vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts
Source: CVE Program / CVE List V5
Vulnerability 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.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N
Source: CVE Program / CVE List V5
Vulnerability Type
使用共享资源的并发执行不恰当同步问题(竞争条件)
Source: CVE Program / CVE List V5
Vulnerability Title
vLLM 竞争条件问题漏洞
Source: CNNVD (China National Vulnerability Database)
Vulnerability Description
vLLM是vLLM团队开源的一个适用于 LLM 的高吞吐量和内存高效推理和服务引擎。 vLLM 0.20.2rc0版本至0.26.0之前版本存在竞争条件问题漏洞，该漏洞源于并发提交的prompt_embeds部分与torch.sparse.check_sparse_tensor_invariants的进程全局状态保存、启用和恢复操作产生竞争，可能导致无效稀疏张量到达tensor.to_dense，绕过CVE-2025-62164防护。
Source: CNNVD (China National Vulnerability Database)
CVSS Information
N/A
Source: CNNVD (China National Vulnerability Database)
Vulnerability Type
N/A
Source: CNNVD (China National Vulnerability Database)

Affected Products

Vendor Product Affected Versions CPE Subscribe
vllm-project vllm >= 0.20.2rc0, < 0.26.0 -

II. Public POCs for CVE-2026-73557

# POC Description Source Link Shenlong Link
AI-Generated POC Premium

No public POC found.

Login to generate AI POC

III. Intelligence Information for CVE-2026-73557

请登录查看更多情报信息。

Patches & Fixes for CVE-2026-73557 (2)

Vendor Advisories for CVE-2026-73557 (1)

Vendor Pages for CVE-2026-73557 (1)

Same Patch Batch · vllm-project · 2026-08-13 · 5 CVEs total

CVE-2026-73559 6.5 MEDIUM vLLM: Completion prompt lists fan out into unbounded engine requests
CVE-2026-73555 5.3 MEDIUM vLLM: Unauthenticated Internal Path and Username Disclosure via Validation Error Messages
CVE-2026-73556 5.3 MEDIUM vLLM: ReDoS via structured_outputs.regex in the lm-format-enforcer backend (no compile tim
CVE-2026-73558 5.3 MEDIUM vLLM: Cross-User Data Leak Vulnerability

IV. Related Vulnerabilities

Same product: vllm
Same vendor: vllm-project
Same weakness: CWE-362

V. Comments for CVE-2026-73557

No comments yet

Leave a comment