vllm-project · vLLM · published 2026-09-26
CVE-2026-100647vLLM before 0.29.0 CPU Exhaustion via unbounded cache_salt
Routine No exploitation evidence; EPSS 0.3%.
CVE-2026-100647 is not known to be exploited in the wild. EPSS estimates a 0.3% chance of exploitation in the next 30 days, higher than 29% of all CVEs. This vLLM vulnerability: CVSS 6.9 (v4.0), attacks can be automated. Fixed upstream in vLLM 0.29.0, 0.28.0, 0.24.0. Routine: patch when convenient.
Jump to: Remediation Exploit intelAffected versionsDescription
Remediation#
Fixed upstream in vLLM 0.29.0, 0.28.0, 0.24.0.
Patch events
AI toolingvLLM
vLLM security releases: 0.29.0, 0.28.0, 0.24.0
Upgrade to
0.29.00.28.00.24.0Why: Vendor security release for server software (most severe: CVE-2026-100647).
Exploit intelligence#
| Signal | Value | Source |
|---|---|---|
| Known exploited | No | CISA KEV |
| Ransomware use | — | CISA KEV |
| EPSS (30-day exploitation probability) | 0.3% (29th pct) | FIRST EPSS v2026.06.15 |
| CVSS base score | 6.9 (v4.0)CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N | CNA / CISA ADP |
| SSVC exploitation | none | CISA Vulnrichment |
| SSVC automatable | Yes | CISA Vulnrichment |
| SSVC technical impact | partial | CISA Vulnrichment |
Affected versions (from the CVE record)#
- vllm-project vllm
0to before0.29.0(affected)0.29.0(unaffected)
Description (from the CVE record)#
vLLM versions before 0.29.0 contain a denial-of-service vulnerability in the cache_salt parameter accepted on OpenAI-compatible and Anthropic API endpoints, which lacks maximum length validation and is processed on the single EngineCore scheduler thread. Unauthenticated attackers can send HTTP requests with multi-hundred-megabyte salt values that trigger expensive pickle serialization and SHA-256 hashing, stalling the scheduler thread and denying service to all concurrent requests.
Quoted from the CVE Program record.