NVD / CVE Program· CNA VulnCheck

CVE-2026-100651: vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass

HighCVSS 7.1 · v4.0—No exploit Fix available

Noroxi assessment

Priority: Monitor (action score 28/100)

The sources we track (CISA KEV, Exploit-DB, Metasploit, Nuclei, GitHub) show no public exploit code or exploitation in the wild.
Probability of exploitation in the next 30 days: 0.38% (FIRST EPSS).
Anyone who can reach it over the internet can trigger it. A low-privileged account is enough. No user action is required. No special conditions are required; it is repeatable.
Fixed version: vllm-project vllm 0.29.0.

Low priority for now. Upgrade to vllm-project vllm 0.29.0 or later.

Generated automatically from the record's data: CVSS (NVD and CNA), CISA KEV and SSVC, FIRST EPSS, exploit sources, fixed versions. How we calculate it

Primary sources: NVD · CVE.org · FIRST EPSS

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.

Denial of service

EUVD-2026-87745

Published: Sep 26, 2026
Updated: Oct 6, 2026
EPSS: 0.3% · 32th percentile
CWE: CWE-400

Report tools

JSON

Action score

Monitor

Low priority for now.

CVSS: 28 / 40 · 7.1 / 10
CISA KEV: 0 / 30 · Not listed
EPSS: 0 / 30 · 0.3%

CISA SSVC decision

Exploitation: proof of concept
Automatable: no
Technical impact: partial

Vulnrichment: CISA's decision-tree inputs.

CNA vs NVD score

NVD: —
CNA · VulnCheck: 7.1; NVD has not scored this yet; the score shown is the CNA’s.

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

The score given by the assigning authority versus NVD’s independent score. A gap means the severity is contested.

Get told when this record changes

Not in KEV: no record of exploitation in the wild
No public exploit code on record
A fix has been released

If any of these changes, we'll tell you in the daily email digest.

Is your version affected? Check it

Noroxi analysis

What does it mean in your environment?

The assessment above is generated automatically from the record's data; it doesn't know your environment. Our team verifies on your systems whether this vulnerability is actually exploitable and which mitigation is enough.

We use this product, ask for help

Affected systems

Is your version affected?

Pick the product and type the version you run. It's compared with this record's version ranges in your browser; the version isn't sent anywhere.

Vendor	Product	CPE
vllm	vllm	cpe:2.3:a:vllm:vllm

Versions reported by the vendor

Affected version ranges reported by the assigning authority (VulnCheck). Independent of NVD's CPE analysis and usually ahead of it.

vllm-project vllm
- 0.29.0not affected
- before 0.29.0affected · semver

Affected versions

NVD version ranges (for catalog products). Add the product to your stack with a version and matching uses these.

vllm vllmbefore 0.29.0

Same product

vllm: all records

Other highest-scoring records for the same primary product.

CVE-2026-22778vLLM leaks a heap address when PIL throws an error
60This week
CVE-2025-32444vLLM Vulnerable to Remote Code Execution via Mooncake Integration
40Plan
CVE-2024-11041Remote Code Execution in vllm-project/vllm
40Plan

Published before and after this one in the same product

← EarlierCVE-2026-100649Sep 26, 2026 · vLLM before 0.29.0 Resource Limit Bypass via Sampler Subclass
25Monitor
← EarlierCVE-2026-100650Sep 26, 2026 · vLLM before 0.29.0 Resource Exhaustion via Unbounded Media Materialization
28Monitor
Later →CVE-2026-100652Sep 26, 2026 · vLLM 0.22.0 through 0.23.0 Denial of Service via stop_token_ids
32Monitor

Remediation

Which version to upgrade to

Fix versions compiled from the vendor, package registries and Microsoft. Verify the vendor's note before upgrading.

Product / package	Fixed version	Source
vllm-project vllm	0.29.0	Vendor (CNA)

Exploit status

No known public exploit

No public exploit has been observed yet. That doesn't mean you're safe, only that the bar is a little higher.

Research context

For pentesters and researchers: attack profile, score disagreement, timeline, patch commits, credits, variant and chain candidates, bug bounty scope. All derived from existing data; no exploit code.

Timeline

From publication to today: proof of concept, Metasploit module, CISA KEV and fix record. Dates are as reported by the sources.

No dated events beyond publication.

EPSS, last 120 days

FIRST EPSS daily score; only changes of 0.01 or more are recorded (step chart).

Credits

All researchers

Finders, reporters and analysts named in the CNA record. Click a name for that researcher’s other records.

rexpositoryreporter
jperezdealgabacoordinator

Variant candidates

Same product, same weakness class, within 18 months. If the patch missed the root cause, the sibling bug is here.

CVE-2026-11405525 days apartvLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs
34Monitor
CVE-2026-11617729 days apartvLLM 0.10.2 before 0.25.0 DoS via Nemotron-VL Processor Decompression-Bomb Guard Bypass
34Monitor
CVE-2026-1057599 days apartvLLM: Unbounded Prometheus label cardinality from attacker-controlled HTTP method tokens in the vLLM Rust frontend metrics middleware (unauthenticated denial…
30Monitor
CVE-2026-3723729 days apartvLLM up to and including 0.17.0 allows remote attackers to cause a Denial of Service via memory exhaustion.
30Monitor
CVE-2026-5497107 days apartUnbounded Frame Count in video/jpeg Base64 Data URL Processing Leads to OOM DoS in vllm-project/vllm
30Monitor
CVE-2025-48956401 days apartvLLM API endpoints vulnerable to Denial of Service Attacks
30Monitor

Not on record for this CVE: patch commit or PR link · public bounty program.

Technical details

Attack conditions

Anyone who can reach it over the internet can trigger it.
A low-privileged account is enough.
No user action is required.
No special conditions are required; it is repeatable.

If successful

Confidentiality
Integrity
Availability

Attack vector: Network
Attack complexity: Low
Privileges required: Low
User interaction: None
Scope: X

Weakness class (CWE)

CWE-400 · Uncontrolled Resource Consumption

CVSS vector

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/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

nvd-secondary

Root cause

Code that parses nested structures sets no depth or size limit. A small input can turn into a disproportionate processing load.

The bug in code

The whole CWE-400 class

A representative example of this vulnerability class. Not the vendor's source code; it shows the faulty pattern and its fix.

Vulnerable

1function expand(node) {
2  return node.children.map(expand);
3}

Fixed

1const MAX_DEPTH = 32;
2function expand(node, depth = 0) {
3  if (depth > MAX_DEPTH) throw new Error("too deep");
4  return node.children.map((c) => expand(c, depth + 1));
5}

Attack context

MITRE CAPEC attack patterns and ATT&CK techniques for this weakness class (CWE). A starting point for detection rules and threat hunting.

Attack patterns (CAPEC)

ATT&CK techniques

Endpoint Denial of ServiceT1499

Change log (2)

Fix✗ → ✓Sep 28, 2026, 8:49 PM UTC
SSVC exploitation— → pocSep 28, 2026, 8:49 PM UTC

For records you follow, these changes also arrive as notifications. →

References

www.vulncheck.com/advisories/vllm-before-0.29.0-denial-of-service-via-decoder-prompt-length-bypass

Vendor advisories and official records. Exploit/PoC links are deliberately left out.

All records