ModelTC vulnerabilities
Published CVEs affecting ModelTC products: 3 record(s), with severity, exploitation and fixes.
CVE-2026-111160
High · 7.5
CWE-502 Oct 15, 2026
LightLLM through 1.2.0 Unauthenticated DoS via pd_kv_trans_params Deserialization
LightLLM through 1.2.0 contains a denial of service vulnerability that allows unauthenticated attackers to crash the inference worker by supplying the internal pd_kv_trans_params field to the /generate endpoint. Attackers can send a single request whose parameters cause pickle.loads to fail in InferSamplingParams, killing the inference thread while /health stays up and all later requests hang until restart.
CVE-2026-108943
Medium · 5.3
CWE-74 Oct 12, 2026
ModelTC LightX2V Tokenizer cosmos3_runner.py AutoTokenizer.from_pretrained code injection
A weakness has been identified in ModelTC LightX2V up to 0.5.0. Affected is the function AutoTokenizer.from_pretrained of the file lightx2v/models/runners/cosmos3/cosmos3_runner.py of the component Tokenizer. This manipulation of the argument auto_map causes code injection. The attack is restricted to local execution. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.
CVE-2026-26220
Critical · 9.3
CWE-502 Feb 17, 2026
LightLLM <= 1.1.0 PD Mode Unsafe Deserialization RCE
LightLLM version 1.1.0 and prior contain an unauthenticated remote code execution vulnerability in PD (prefill-decode) disaggregation mode. The PD master node exposes WebSocket endpoints that receive binary frames and pass the data directly to pickle.loads() without authentication or validation. A remote attacker who can reach the PD master can send a crafted payload to achieve arbitrary code execution.