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

Goal: 1000 CNY · Raised: 1359 CNY

100%
Buy Us a Coffee

CVE-2026-100654— vLLM before 0.29.0 Denial of Service via out-of-range stop_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 之前的版本中，系统在 OpenAI 兼容的 POST /v1/completions 和 POST /v1/chat/completions 端点上接受用户控制的 stop_token_ids 参数，但仅验证这些值是否为整数，而未检查每个 token ID 是否位于模型词汇表或 logits 索引的有效范围内。 当 min_tokens 参数大于 0 时，stop token IDs 会被用作 logits 索引，以抑制停止令牌。此时，若传入超出有效范围的 token ID，将触发 CU

CVSS 6.5 · Medium EPSS 0.31% · P21
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-100654

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Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.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 Denial of Service via out-of-range stop_token_ids
Source: CVE Program / CVE List V5
Vulnerability Description
vLLM before 0.29.0 accepts user-controlled stop_token_ids on the OpenAI-compatible POST /v1/completions and POST /v1/chat/completions endpoints but validates only that the values are integers, not that each token id is within the model vocabulary/logits range. When min_tokens > 0, the stop token ids are used as logits indices to suppress stop tokens, so an out-of-range id reaches a CUDA indexing operation (index_put_) and triggers a device-side assertion. An authenticated API user can send a single malformed completion request that returns 500 Internal Server Error and puts EngineCore into a fatal state, causing subsequent requests to fail until the service is restarted (denial of service).
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
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-2026-100654

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

No public POC found.

Login to generate AI POC

III. Intelligence Information for CVE-2026-100654

请登录查看更多情报信息。

Vendor Advisories for CVE-2026-100654 (1)

Other References for CVE-2026-100654 (1)

Same Patch Batch · vllm-project · 2026-09-26 · 8 CVEs total

CVE-2026-100650 6.5 MEDIUM vLLM before 0.29.0 Resource Exhaustion via Unbounded Media Materialization
CVE-2026-100651 6.5 MEDIUM vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass
CVE-2026-100653 6.5 MEDIUM vLLM 0.22.1 before 0.28.0 Incomplete Artifact Pin Propagation
CVE-2026-100652 5.9 MEDIUM vLLM 0.22.0 through 0.23.0 Denial of Service via stop_token_ids
CVE-2026-100647 5.3 MEDIUM vLLM before 0.29.0 CPU Exhaustion via unbounded cache_salt
CVE-2026-100648 5.3 MEDIUM vllm before 0.29.0 Uncontrolled Resource Consumption via Audio Decoding
CVE-2026-100649 3.7 LOW vLLM before 0.29.0 Resource Limit Bypass via Sampler Subclass

IV. Related Vulnerabilities

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

V. Comments for CVE-2026-100654

No comments yet

Leave a comment