Skip to content
Noroxi
NVD / CVE Program· CNA GitHub_M

CVE-2026-54234: vLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection

HighCVSS 7.5 · v3.1—No exploit Fix available

Noroxi assessment

Priority: Monitor (action score 30/100)

  • 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: 0.61% (FIRST EPSS); higher than 50.1% of all records.
  • Anyone who can reach it over the internet can trigger it. No account or password is required. No user action is required. No special conditions are required; it is repeatable.
  • According to CISA the attack is automatable: targets can be found by mass scanning.
  • Fixed version: vllm-project vllm 0.24.0.

Low priority for now. Upgrade to vllm-project vllm 0.24.0 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 is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to 0.24.0, a frontend-legal multi-request speculative decoding workload can cause the rejection sampler to produce a recovered token equal to the model vocabulary size boundary value, which is then converted to negative one when the engine selects the next live token for a request and is written back into the drafter's input ids; that out-of-vocabulary value is later consumed by the model's embedding and attention path and crashes the engine worker with a GPU device-side assertion. The same triggering request sequence is reachable through the public gRPC Generate and Abort endpoints, so a remote client that can send generation requests can crash the shared engine worker, aborting concurrent requests and causing a service-wide denial of service for other clients of the deployment until the worker is restarted. This issue is fixed in version 0.24.0.

Published
Jul 6, 2026
Updated
Oct 25, 2026
EPSS
0.6% · 50th percentile

Report tools

JSON

Action score

30

Monitor

Low priority for now.

CVSS
30 / 40 · 7.5 / 10
CISA KEV
0 / 30 · Not listed
EPSS
0 / 30 · 0.6%

CISA SSVC decision

Exploitation
proof of concept
Automatable
yes
Technical impact
partial

Vulnrichment: CISA's decision-tree inputs.

CNA vs NVD score

NVD
—
CNA · GitHub_M
7.5
NVD has not scored this yet; the score shown is the CNA’s.

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.

Follow CVE-2026-54234

When you'll hear from us

  • when it enters CISA KEV
  • when exploit code appears or is weaponized
  • when the record is updated (including a fixed version)
  • when EPSS rises 20 points in a week

One daily email digest (06:00 UTC); connect a webhook for KEV listings and weaponized exploits every 10 minutes on paid plans, on the hour on the free plan.

Your free account also gets you

  • CVE alerts: add the products and versions you run, get the new CVEs that affect them
  • Uptime monitoring: 1 site, instant email when it goes down
  • Add your domain: subdomains and takeover risk scanned every week (1 domain)
  • 10 scans an hour on the security tools, at most 30 a day (3 a day without an account), an API key and MCP
  • All your scans in one panel: findings, what changed in the last scan, a report per domain or per client (PDF, share link)
All free tools
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.

VendorProduct
vllmvllm

Versions reported by the vendor

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

  • vllm-project vllm

    • < 0.24.0affected

Affected versions

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

  • vllm vllmbefore 0.24.0

Package-level exposure

OSV and GitHub Advisory data: ecosystem, package and range. SBOM matching uses this table.

EcosystemPackageAffected rangeFix
PyPIvllmfrom 0.17.1 · before 0.24.00.24.0

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-2026-54236Jun 22, 2026 · vLLM: incomplete CVE-2026-22778 fix leaks PIL repr addresses via Anthropic router
    21Monitor
  • ← EarlierCVE-2026-55646Jul 6, 2026 · vLLM speech-to-text endpoints allocate full upload before enforcing the audio file-size limit
    26Monitor
  • Later →CVE-2026-55514Jul 6, 2026 · vLLM denial of service via prompt embeds on M-RoPE models
    28Monitor

Remediation

Which version to upgrade to

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

Product / packageFixed versionSource
vllm-project vllm0.24.0 · descriptionVendor (CNA)
PyPI:vllm0.24.0Package registry (OSV)

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.

Patch and commit links

Commit, PR and diff links among the references. A starting point for patch-diffing and variant hunting; fixes, not exploits.

Variant candidates

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

  • CVE-2026-115915112 days apartvLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCE
    39Monitor
  • CVE-2025-62164227 days apartVLLM deserialization vulnerability leading to DoS and potential RCE
    35Monitor
  • CVE-2026-5634016 days apartvLLM - Denial of Service via Unvalidated Multimodal Embeddings
    34Monitor
  • CVE-2026-10065282 days apartvLLM 0.22.0 through 0.23.0 Denial of Service via stop_token_ids
    32Monitor
  • CVE-2026-116172111 days apartvLLM 0.22.0 before 0.24.0 DoS via Rust gRPC CandidateTokens.top_n
    32Monitor
  • CVE-2026-116178111 days apartvLLM 0.20.0 before 0.30.0 DoS via /inference/v1/generate mm_placeholders
    32Monitor

Not on record for this CVE: EPSS history · researcher credit · public bounty program.

Technical details

Attack conditions

  • Anyone who can reach it over the internet can trigger it.
  • No account or password is required.
  • No user action is required.
  • No special conditions are required; it is repeatable.

If successful

Confidentiality
none
Integrity
none
Availability
high · the service can be disrupted
Attack vector
Network
Attack complexity
Low
Privileges required
None
User interaction
None
Scope
Unchanged
Confidentiality impact
None
Integrity impact
None
Availability impact
High

CVSS vector

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

nvd-secondary

References

Show all references (1 more)

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

All records