Technology · ModelTC
ModelTC LightLLM vulnerabilities
Updated . Rebuilt every hour.
Junglewise Threat Intelligence has tracked 11 vulnerabilities in ModelTC LightLLM: 0 in the last 7 days and 10 in the last 90 days, 6 of them critical and 0 exploited in the wild. The most recent, CVE-2026-111160, was published on 15 October 2026. 1 vulnerability in the last four weeks, down 89% on the four before (9). 0 critical against 6.
- Last 7 days
- 0
- Last 90 days
- 10
- Critical, all time
- 6
- Exploited in the wild
- 0
About ModelTC LightLLM
A lightweight Large Language Model (LLM) inference framework designed for high-throughput serving.
Latest ModelTC LightLLM vulnerabilities
- CVE-2026-111160: LightLLM denial of service via pd_kv_trans_params unpicklinghighCVSS 7.5EPSS 0.4%
- CVE-2026-103395: LightLLM visual_only unauthenticated RCE via RPyC pickle deserializationcriticalCVSS 9.8
- CVE-2026-103270: LightLLM authentication bypass in reinforcement learning control APIhighCVSS 7.5
- CVE-2026-103243: LightLLM server-side request forgery in multimodal endpointsmediumCVSS 5.8
- CVE-2026-103042: LightLLM memory exhaustion in NCCL control channelhighCVSS 7.5
- CVE-2026-103041: LightLLM multimodal pickle deserialization in RPyC cache servicecriticalCVSS 9.8
- CVE-2026-103040: LightLLM remote code execution in router profiler servicecriticalCVSS 9.8
- CVE-2026-96560: LightLLM KV-transfer worker remote code execution in NCCL modecriticalCVSS 9.8EPSS 0.7%
- CVE-2026-93839: LightLLM authentication bypass in /pd_register WebSocket endpointcriticalCVSS 9.8EPSS 1.0%
- CVE-2026-90919: LightLLM Config Server unsafe pickle deserialization in /visual_registercriticalCVSS 9.8EPSS 1.1%
- CVE-2026-26220: ModelTC LightLLM remote code execution in PD WebSocket endpointsinfoCVSS 9.3EPSS 0.7%
Most severe ModelTC LightLLM vulnerabilities
Exploited in the wild first, then by severity and CVSS score.
- CVE-2026-90919: LightLLM Config Server unsafe pickle deserialization in /visual_registercriticalCVSS 9.8EPSS 1.1%
- CVE-2026-93839: LightLLM authentication bypass in /pd_register WebSocket endpointcriticalCVSS 9.8EPSS 1.0%
- CVE-2026-96560: LightLLM KV-transfer worker remote code execution in NCCL modecriticalCVSS 9.8EPSS 0.7%
- CVE-2026-103395: LightLLM visual_only unauthenticated RCE via RPyC pickle deserializationcriticalCVSS 9.8
- CVE-2026-103041: LightLLM multimodal pickle deserialization in RPyC cache servicecriticalCVSS 9.8
- CVE-2026-103040: LightLLM remote code execution in router profiler servicecriticalCVSS 9.8
- CVE-2026-111160: LightLLM denial of service via pd_kv_trans_params unpicklinghighCVSS 7.5EPSS 0.4%
- CVE-2026-103270: LightLLM authentication bypass in reinforcement learning control APIhighCVSS 7.5
- CVE-2026-103042: LightLLM memory exhaustion in NCCL control channelhighCVSS 7.5
- CVE-2026-103243: LightLLM server-side request forgery in multimodal endpointsmediumCVSS 5.8
Vulnerabilities per week
1 vulnerability in the last four weeks, down 89% on the four before (9). 0 critical against 6. The last 13 weeks, by the week each vulnerability was published.
| Week of | Bar | Vulns | Critical |
|---|---|---|---|
| 17 Aug 2026 | 0 | 0 | |
| 24 Aug 2026 | 0 | 0 | |
| 31 Aug 2026 | 0 | 0 | |
| 7 Sep 2026 | 0 | 0 | |
| 14 Sep 2026 | 2 | 2 | |
| 21 Sep 2026 | 1 | 1 | |
| 28 Sep 2026 | 6 | 3 | |
| 5 Oct 2026 | 0 | 0 | |
| 12 Oct 2026 | 1 | 0 | |
| 19 Oct 2026 | 0 | 0 | |
| 26 Oct 2026 | 0 | 0 | |
| 2 Nov 2026 | 0 | 0 | |
| 9 Nov 2026 | 0 | 0 |
Embed a badge
A live badge for a README, a status page or an internal wiki: how many ModelTC LightLLM vulnerabilities were published in the last 7 days and how many of them are critical, refreshed hourly.
[](https://junglewise.ai/threats/technologies/lightllm)<a href="https://junglewise.ai/threats/technologies/lightllm"><img src="https://junglewise.ai/threats/technologies/lightllm/badge.svg" alt="ModelTC LightLLM vulnerabilities" /></a>How this is built
Junglewise Threat Intelligence collects vulnerabilities from NVD, GitHub Security Advisories, OSV, the CISA Known Exploited Vulnerabilities catalog, FIRST EPSS and vendor advisories, and matches each one to the technologies and vendors it affects. Dates are the date a vulnerability was published, in UTC.
The pages are rebuilt from the database every hour. Frozen weekly and monthly reports never change once published, so they can be cited.
Use this data
The same data is at https://junglewise.ai/threats/technologies/lightllm.json, and this page as text at https://junglewise.ai/threats/technologies/lightllm.md, for scripts and language models. It is free to reuse under CC BY 4.0 with a link back to this page.
Cite as: Junglewise Threat Intelligence, "ModelTC LightLLM vulnerabilities", https://junglewise.ai/threats/technologies/lightllm, 9 November 2026.