CVE-2026-76850

LMDeploy deserializes disaggregated-serving peer messages with pickle. The handlezmqrecv coroutine in lmdeploy/pytorch/disagg/conn/engineconn.py reads peer-to-peer cache-free requests with recvpyobj, which deserializes the received bytes with pickle.loads, and the isinstance check against...

CVE-2026-76850

LMDeploy (InternLM) is vulnerable to remote code execution via unsafe pickle deserialization in its disaggregated-serving peer connector. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py deserializes peer-to-peer messages with pickle.loads() before any type validation,...

The vulnerability of the recv_pyobj method in the Llama Stack framework, which is used for working with large language models (LLMs), allows a attacker to execute arbitrary code.

The vulnerability of the recvpyobj method in the Llama Stack framework for working with large language models is related to deficiencies in the deserialization mechanism. Exploiting this vulnerability could allow a malicious actor to execute arbitrary code using a specially crafted JSON file...