CVE-2026-103395

LightLLM through 1.2.0 visualonly deployments expose an unauthenticated RPyC service with allowpickle enabled that deserializes attacker-supplied arguments in the remoteinferimages method. Attackers can reach the visual RPyC port and pass objects with reduce methods to execute arbitrary code with...

CVE-2026-103270

LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pausegeneration, /abortrequest, /flushcache, and /initweightsupdategroup to disrupt inference operations and wedge workers o...

CVE-2026-103243

LightLLM through 1.2.0 fails to validate imageurl and audiourl parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error...

CVE-2026-103270 LightLLM through 1.2.0 Missing Authentication on RL Control Routes

LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pausegeneration, /abortrequest, /flushcache, and /initweightsupdategroup to disrupt inference operations and wedge workers o...

EUVD-2026-89989

LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pausegeneration, /abortrequest, /flushcache, and /initweightsupdategroup to disrupt inference operations and wedge workers o...

CVE-2026-103270 LightLLM through 1.2.0 Missing Authentication on RL Control Routes

LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pausegeneration, /abortrequest, /flushcache, and /initweightsupdategroup to disrupt inference operations and wedge workers o...

EUVD-2026-89990

LightLLM through 1.2.0 visualonly deployments expose an unauthenticated RPyC service with allowpickle enabled that deserializes attacker-supplied arguments in the remoteinferimages method. Attackers can reach the visual RPyC port and pass objects with reduce methods to execute arbitrary code with...

CVE-2026-103395 LightLLM through 1.2.0 Unauthenticated Remote Code Execution via Visual-Only RPyC Service

LightLLM through 1.2.0 visualonly deployments expose an unauthenticated RPyC service with allowpickle enabled that deserializes attacker-supplied arguments in the remoteinferimages method. Attackers can reach the visual RPyC port and pass objects with reduce methods to execute arbitrary code with...

CVE-2026-103395 LightLLM through 1.2.0 Unauthenticated Remote Code Execution via Visual-Only RPyC Service

LightLLM through 1.2.0 visualonly deployments expose an unauthenticated RPyC service with allowpickle enabled that deserializes attacker-supplied arguments in the remoteinferimages method. Attackers can reach the visual RPyC port and pass objects with reduce methods to execute arbitrary code with...

CVE-2026-103395

Versions of LightLLM through 1.2.0 with visual_only deployments are vulnerable to Unauthenticated Remote Code Execution (RCE) . The issue stems from an exposed RPyC service where the allow_pickle option is enabled, allowing the remote_infer_images method to deserialize untrusted arguments. An att...

CVE-2026-103270

LightLLM versions up to and including 1.2.0 are vulnerable to a missing authentication flaw on reinforcement learning (RL) control routes. When deployments are started with the --enable_rl flag, the software mounts specific control routes on the public HTTP API without requiring authentication ch...

CVE-2026-103395 LightLLM through 1.2.0 Unauthenticated Remote Code Execution via Visual-Only RPyC Service

LightLLM through 1.2.0 visualonly deployments expose an unauthenticated RPyC service with allowpickle enabled that deserializes attacker-supplied arguments in the remoteinferimages method. Attackers can reach the visual RPyC port and pass objects with reduce methods to execute arbitrary code with...

EUVD-2026-89988

LightLLM through 1.2.0 fails to validate imageurl and audiourl parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error...

CVE-2026-103243 LightLLM through 1.2.0 Server-Side Request Forgery via multimodal endpoints

LightLLM through 1.2.0 fails to validate imageurl and audiourl parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error...

CVE-2026-103243 LightLLM through 1.2.0 Server-Side Request Forgery via multimodal endpoints

LightLLM through 1.2.0 fails to validate imageurl and audiourl parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error...

CVE-2026-103243

Versions of LightLLM up to and including 1.2.0 are vulnerable to a Server-Side Request Forgery (SSRF) . The root cause is a failure to validate the image_url and audio_url parameters within multimodal endpoints . This allows unauthenticated attackers to supply arbitrary URLs to fetch internal res...

CVE-2026-103243 LightLLM through 1.2.0 Server-Side Request Forgery via multimodal endpoints

LightLLM through 1.2.0 fails to validate imageurl and audiourl parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error...

EUVD-2026-89483

LightLLM through 1.2.0 contains a memory exhaustion vulnerability in the NCCL control channel when started with --pdtransmode nccl, allowing unauthenticated attackers to exhaust KV-transfer worker memory. Attackers can call the exposedsetvalue method to store unbounded key-value pairs without siz...

EUVD-2026-89482

LightLLM through 1.2.0 multimodal deployments expose an unauthenticated RPyC cache service with pickle deserialization enabled on all interfaces. Attackers can send crafted serialized objects to exposed cache methods to execute arbitrary code with service privileges...

EUVD-2026-89481

LightLLM through 1.2.0 contains a remote code execution vulnerability in the router profiler service when started with --enableprofiling flag. The service exposes an unauthenticated RPyC server with pickle deserialization enabled, allowing attackers to execute arbitrary code by sending crafted...