CVE Catalog

CVE-2026-103270

HighCVSS 7.5
Published: Updated: Translated: NVD NIST

Summary

LightLLM through version 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.

Risk Assessment

An attacker can disrupt inference services, potentially leading to availability outages and system failures.

Recommendation

Upgrade LightLLM to a version newer than 1.2.0 that includes authentication fixes for these routes.

Other vulnerabilities in LightLLM

See all
Original NVD description (English source)

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.

Vulnerability data from NVD (NIST) · CISA KEV · EPSS