MediumVulnerabilityLLM-specific

CVE-2026-126398: vLLM reachable assertion crash in GPUModelRunner via embeddings-only requests

Source: NVD(opens in a new tab)
Identifier: CVE-2026-126398
Published: November 13, 2026
Record updated: November 13, 2026

Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 0.5%

Summary

vLLM versions from 0.28.0 before 0.30.0 contain a reachable assertion in GPUModelRunner._init_xdrope_positions. A remote attacker can send a /v1/completions request containing only prompt_embeds to XD-RoPE models run with --enable-prompt-embeds, which terminates the EngineCore and API server for all clients.

Mitigation

Fixed in 0.30.0 (upgrade vLLM to 0.30.0 or later).

Details

Scores and exploitation

CVSS score: 5.9(medium)
CVSS vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H
Attack vector: network
Attack complexity: high
Privileges required: none
User interaction: none
EPSS: 0.5%Estimated probability that this CVE is exploited in the next 30 days, from the FIRST Exploit Prediction Scoring System. Read on November 13, 2026.
Disclosure date: November 13, 2026

Weaknesses and techniques

Weakness (CWE): CWE-617

Classification

Attack type: Denial of service
Attack sophistication: Trivial
AI component targeted: Inference
Impact: Availability
Labels: security

How this record was produced

Summaries, mitigations and labels on this page are generated automatically from the source. The linked source is authoritative. The source gives this advisory a description and no title, so the title above is a headline generated from the source text.

Original source: https://nvd.nist.gov/vuln/detail/CVE-2026-126398

First tracked: November 13, 2026 at 01:10 PM ET, from NVD/CVE Database

Classifier: claude-haiku-5-5, prompt v4

Classifier confidence: 95% (self-assessed)

Topics

Inference infrastructure

Summary

Mitigation

Topics

Related items