Public mode

No access to your tenant or devices. Every item links to its source. What this means

Sources need attention Customize Public modeDiscuss ChangeIntel on Discord

vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass

CVSS 4.0 7.1 HIGHCVSS 3.1 6.5 MEDIUMCheck source for fixed versionsKEV status unknown

Historical advisory recorded 23 Oct 2026 15:48:27 UTC. Read the latest advisory.

cve-program · CVE-2026-100651

Published

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' (multimodal) payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied token_ids, and GenerateRequest.token_ids (vllm/entrypoints/serve/disagg/protocol.py) is not checked against model_config.max_model_len. For multimodal processors that report skip_prompt_length_check=True (for example Nemotron Parse, Whisper, and FireRedLID), InputProcessor._validate_prompt_len() returns immediately for both encoder and decoder prompts, so an overlong prompt becomes an EngineCoreRequest and reaches the worker input-batch copy into a fixed max_model_len-wide NumPy row. A client able to reach the endpoint on an affected model configuration can therefore submit an overlong token_ids list to trigger a worker failure and denial of service. Fixed in 0.29.0.

Read the publisher's advisory

Affected versions, risk and source history

cve-program · CVE-2026-100651

Published

vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' (multimodal) payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied token_ids, and GenerateRequest.token_ids (vllm/entrypoints/serve/disagg/protocol.py) is not checked against model_config.max_model_len. For multimodal processors that report skip_prompt_length_check=True (for example Nemotron Parse, Whisper, and FireRedLID), InputProcessor._validate_prompt_len() returns immediately for both encoder and decoder prompts, so an overlong prompt becomes an EngineCoreRequest and reaches the worker input-batch copy into a fixed max_model_len-wide NumPy row. A client able to reach the endpoint on an affected model configuration can therefore submit an overlong token_ids list to trigger a worker failure and denial of service. Fixed in 0.29.0.

Source published
26 Sep 2026 13:23:21 UTC
Source modified
28 Sep 2026 18:25:40 UTC
Saved here
23 Oct 2026 15:48:27 UTC
Source terms
CVE Program Terms of Use

Original source record · Cite this exact observation · Complete original publisher JSON

Source: cve-program · CVE Program Terms of Use. The publisher's wording is preserved. ChangeIntel groups the information and labels changes. The JSON record includes additional source fields.

CVSS 4.0 · 7.1 CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N

CVSS 3.1 · 6.5 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

ChangeIntel

An IT change radar: releases, security, known issues, retirements, documentation changes, and service status from public sources. Every item links to supporting evidence; dates and statuses can change after they are read.

Sources read 16 Nov 08:58 UTC · 274 of 275 readable · 252 fully read · 18 waiting for GitHub's hourly limit · documentation 469/469 current