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

Goal: 1000 CNY · Raised: 1336 CNY

100%
Buy Us a Coffee

CVE-2026-93840— vLLM before 0.29.0 Cross-Request Logits Corruption via allowed_token_ids

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.

在 vLLM 0.29.0 之前， 函数在验证 时，依据的是分词器（tokenizer）的长度，而非模型输出 logits 的宽度。攻击者可以传入超过输出词汇表范围的 token ID，使其通过验证，从而导致 损坏 GPU logits 状态，并使得并发请求能够采样出不在其允许列表（allowlist）中的 token。

CVSS 3.7 · Low

Possible ATT&CK Techniques 1 AI

T1204 · User Execution

Affected Version Matrix 2

VendorProduct Version RangeStatus
vllm-project vllm < 0.29.0 affected
0.29.0 unaffected
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-93840

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Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.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 before 0.29.0 Cross-Request Logits Corruption via allowed_token_ids
Source: CVE Program / CVE List V5
Vulnerability Description
vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). Attackers can supply token IDs above the output vocabulary that pass validation, causing LogitBiasState to corrupt GPU logits state and allow concurrent requests to sample tokens outside their allowlists.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N
Source: CVE Program / CVE List V5
Vulnerability Type
对数组索引的验证不恰当
Source: CVE Program / CVE List V5

Affected Products

Vendor Product Affected Versions CPE Subscribe
vllm-project vllm 0 ~ 0.29.0 -

II. Public POCs for CVE-2026-93840

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

No public POC found.

Login to generate AI POC

III. Intelligence Information for CVE-2026-93840

请登录查看更多情报信息。

Patches & Fixes for CVE-2026-93840 (2)

Other References for CVE-2026-93840 (3)

Same Patch Batch · vllm-project · 2026-09-18 · 3 CVEs total

CVE-2026-93592 7.5 HIGH vLLM before 0.28.0 Denial of Service via negative token ID
CVE-2026-93841 3.7 LOW vLLM through 0.29.0 Adjacent-Request Sampler State Corruption via Unvalidated Prompt Token

IV. Related Vulnerabilities

Same product: vllm
Same vendor: vllm-project
Same weakness: CWE-129

V. Comments for CVE-2026-93840

No comments yet

Leave a comment