CVE-2025-48956: vLLM API Denial of Service via Large HTTP Headers

# CVE-2025-48956: vLLM API Denial of Service via Large HTTP Headers

https://gcdradar.vercel.app/incidents/cve-2025-48956-vllm-api-denial-of-service-via-large-http-hea · AIR-2025-0327 · closed 2025-08-21

- Lane: Patch now
- Severity: 2 (Disclosed vulnerability)
- Fix status: Fix status unknown
- First reported: 2025-08-21
- AI role: Attack on AI
- Products: vLLM
- CVE ids: CVE-2025-48956
- CVSS: 7.5

## Summary

An unauthenticated denial of service vulnerability in vLLM occurs when processing excessively large HTTP request headers, leading to server memory exhaustion. Vendor updates are available to resolve the problem.

## What to do

No action recorded; read the sources below and the vendor's advisory.

## Latest updates

- 2025-08-21: CVE-2025-48956: vLLM is an inference and serving engine for large language models (LLMs). (https://nvd.nist.gov/vuln/detail/CVE-2025-48956)
- 2025-08-21: CVE-2025-48956: vllm API endpoints vulnerable to Denial of Service Attacks (https://github.com/advisories/GHSA-rxc4-3w6r-4v47)

## Sources

- 2025-08-21 GitHub Advisory Database: CVE-2025-48956: vllm API endpoints vulnerable to Denial of Service Attacks https://github.com/advisories/GHSA-rxc4-3w6r-4v47
- 2025-08-21 NVD: CVE-2025-48956: vLLM is an inference and serving engine for large language models (LLMs). https://nvd.nist.gov/vuln/detail/CVE-2025-48956

Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.