CVE-2026-103270 LightLLM through 1.2.0 Missing Authentication on RL Control Routes
Unauthenticated attackers can disrupt inference operations and wedge workers on LightLLM deployments if reinforcement learning features are enabled.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- None listed
What is CVE-2026-103270?
CVE-2026-103270 is a high-severity missing authorization vulnerability in ModelTC LightLLM, published on 30 Sep 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.
Official description
LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pause_generation, /abort_request, /flush_cache, and /init_weights_update_group to disrupt inference operations and wedge workers on deployments started with --enable_rl.
Weakness: CWE-306
Which versions are affected by CVE-2026-103270?
How do I fix CVE-2026-103270?
No fix has been published for CVE-2026-103270 yet. Limit exposure of the affected software until one is available.
Upgrade, and check permissions on every request on the server, not only in the interface. More about Missing authorization.
Is CVE-2026-103270 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-103270 exists.
Sources
Also known as GHSA-x6wh-r9r9-mp7w.