Security Digest · Vulnerability record
CVE-2024-9053: vLLM allows Remote Code Execution by Pickle Deserialization via AsyncEngineRPCServer() RPC server entrypoints
vllm-project vllm version 0.6.0 contains a vulnerability in the AsyncEngineRPCServer() RPC server entrypoints.
CVSS 9.8 Critical EPSS 1.4% · p72
Should you act?
Critical and remotely reachable
No known exploitation yet, but it is network-reachable without credentials. This signal is computed from CISA KEV, EPSS and CVSS data; it is not an editorial verdict.
- Not in CISA KEV, but CISA’s SSVC assessment records a public proof-of-concept, so attack code exists.
- EPSS 1.4%, percentile 72: FIRST’s model estimates a 1.4% chance of exploitation activity in the next 30 days, higher than 72% of all scored CVEs (moderate). EPSS is a probability, not a verdict.
- Remotely reachable with no login and no user interaction. If the vulnerable service is exposed to the internet, prioritize it.
- Only applies if your dependency tree includes
vllm(PyPI) <= 0.6.0, no fixed version listed. Check your lockfile, including transitive dependencies.
What it is
vllm-project vllm version 0.6.0 contains a vulnerability in the AsyncEngineRPCServer() RPC server entrypoints. The core functionality runserverloop() calls the function makehandlercoro(), which directly uses cloudpickle.loads() on received messages without any sanitization. This can result in remote code execution by deserializing malicious pickle data.
From the GitHub advisory
vllm-project vllm version 0.6.0 contains a vulnerability in the AsyncEngineRPCServer() RPC server entrypoints. The core functionality runserverloop() calls the function makehandlercoro(), which directly uses cloudpickle.loads() on received messages without any sanitization. This can result in remote code execution by deserializing malicious pickle data.
Weakness type: CWE-502, CWE-78.
Product: vllm (PyPI) · Assigned by: @huntr_ai · Published: 2025-03-20
Affected and fixed versions
| Ecosystem | Package | Vulnerable | First fixed |
|---|---|---|---|
| PyPI | vllm | <= 0.6.0 | none listed |
| Vendor | Product | Affected versions (per CVE record) |
|---|---|---|
| vllm-project | vllm-project/vllm | unspecified <= latest |
Severity scores
| Scored by | CVSS | Score | Vector |
|---|---|---|---|
| @huntr_ai | 3.0 | 9.8 Critical | CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H |
| GitHub | 3.1 | 9.8 | CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H |
- Attack vector: network (remotely reachable)
- Attack complexity: low
- Privileges required: none (no login)
- User interaction: none
- Scope: unchanged
CVSS measures how bad a flaw is if exploited, not how likely exploitation is. Read it next to the exploitation data below.
Exploitation and urgency
- CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
- CISA SSVC assessment: exploitation poc, automatable yes, technical impact total.
- FIRST EPSS (model of 2026-10-21): 1.38% probability of exploitation activity in the next 30 days; percentile 72.
Timeline
- — CVE record published by @huntr_ai
- — GitHub-reviewed advisory GHSA-cj47-qj6g-x7r4 published (critical)
- — CVE record last updated
Sources and references
Original sources first. Version and fix details should always be confirmed against the vendor or project advisory.
Vulnerability records
- GitHub Advisory GHSA-cj47-qj6g-x7r4 github.com
- CVE.org record CVE-2024-9053 cve.org
- NVD entry CVE-2024-9053 nvd.nist.gov
Other references from the CVE record and advisory
- huntr.com huntr.com
- github.com github.com