InternLM LMDeploy

• CVE-2026-119706 Critical 9.8 LMDeploy 0.9.0 through 0.18.0 Unauthenticated RCE via /update_weights Pickle Deserialization 1.0%
• CVE-2026-33625 High 8.8 LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading 0.4%
• CVE-2025-66455 Critical 9.8 LMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py 1.3%
• CVE-2026-92983 High 7.5 InternLM LMDeploy through 0.17.0 Memory Exhaustion via Session ID Mismatch 0.7%
• CVE-2026-92971 High 7.5 InternLM LMDeploy through 0.17.0 Assertion Denial of Service 0.7%
• CVE-2025-59953 Critical 9.8 LMdeploy has Remote Code Execution by Pickle Deserialization via zmq_rpc.call_and_response() in InterLM/lmdeploy 0.9%
• CVE-2026-76850 Critical 9.8 LMDeploy Remote Code Execution via Unsafe Pickle Deserialization in the Disaggregated Serving Peer Connector 1.3%
• CVE-2026-63764 High 8.6 LMDeploy Server-Side Request Forgery via HTTP Redirect Bypass 0.5%
• CVE-2026-46517 High 7.8 LMDeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out 0.2%
• CVE-2026-46432 High 7.8 LMDeploy: Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initialization 0.1%
• CVE-2026-33626 High 7.5 LMDeploy Vulnerable to Server-Side Request Forgery (SSRF) via Vision-Language Image Loading 45.2%
• CVE-2025-67729 High 8.8 lmdeploy vulnerable to Arbitrary Code Execution via Insecure Deserialization in torch.load() 0.5%
• CVE-2025-3163 Medium 5.3 InternLM LMDeploy conf.py open code injection 0.4%
• CVE-2025-3162 Medium 5.3 InternLM LMDeploy PT File utils.py load_weight_ckpt deserialization 0.3%