Vulnerability Description
vLLM is an inference and serving engine for large language models (LLMs). Prior to version 0.9.0, when a new prompt is processed, if the PageAttention mechanism finds a matching prefix chunk, the prefill process speeds up, which is reflected in the TTFT (Time to First Token). These timing differences caused by matching chunks are significant enough to be recognized and exploited. This issue has been patched in version 0.9.0.
CVSS Score
LOW
Affected Products
| Vendor | Product | Versions |
|---|---|---|
| Vllm | Vllm | < 0.9.0 |
Related Weaknesses (CWE)
References
- https://github.com/vllm-project/vllm/commit/77073c77bc2006eb80ea6d5128f076f5e6c6Patch
- https://github.com/vllm-project/vllm/pull/17045Issue TrackingVendor Advisory
- https://github.com/vllm-project/vllm/security/advisories/GHSA-4qjh-9fv9-r85rVendor Advisory
FAQ
What is CVE-2025-46570?
CVE-2025-46570 is a vulnerability with a CVSS score of 2.6 (LOW). vLLM is an inference and serving engine for large language models (LLMs). Prior to version 0.9.0, when a new prompt is processed, if the PageAttention mechanism finds a matching prefix chunk, the pref...
How severe is CVE-2025-46570?
CVE-2025-46570 has been rated LOW with a CVSS base score of 2.6/10. Review the CVSS metrics above for detailed severity breakdown.
Is there a patch for CVE-2025-46570?
Check the references section above for vendor advisories and patch information. Affected products include: Vllm Vllm.