CVE-2026-103270 LightLLM - Broken Access Control

LightLLM through 1.2.0 contains a broken access control vulnerability caused by reinforcement learning control routes on the public HTTP API lacking authentication checks, letting unauthenticated attackers disrupt inference operations, exploit requires deployments started with --enable_rl.

CVSS base score 7.5 /10
High
CISA KEV
Not listed
Public exploit
None known
Authentication
Not required
EPSS
0.51% chance of exploitation in 30 days
View the NVD record ↗

Details

Type
Broken Access Control
Impact
Denial Of Service
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Published
30 September 2026
Updated
15 October 2026
Weakness
CWE-306

Vendor fix

Update to the latest version with authentication checks on reinforcement learning control routes.

Patching closes this CVE. It does not tell you which of your systems run the affected version, or what else is exposed alongside them.

References

← All CVEs