Live Exposure Defense: From CVE to Confirmed Exposure in 12 Hours – See more

Back
New CVE Detected

CVE-2026-90919 – Unauthenticated RCE via Pickle Deserialization – LightLLM Config Server (through 1.

Be the first to know when new zero-days emerge:

Subscribe to Email Alerts
Subscribe to RSS Feed

Summary

CVE-2026-90919 is a critical unauthenticated remote code execution vulnerability in LightLLM, an open-source LLM inference serving framework maintained by ModelTC. The flaw exists in the Config Server component’s /visual_register WebSocket endpoint, which passes the first data frame received from a client directly into Python’s pickle.loads() without any authentication, validation, or filtering. Because pickle deserialization can trigger arbitrary object-reconstruction logic (via __reduce__), a remote, unauthenticated attacker can send a crafted payload to execute arbitrary code under the privileges of the Config Server process.

Technical details

  • Root cause: The visual_websocket_endpoint handler in lightllm/server/config_server/api_http.py calls pickle.loads() directly on the raw bytes of the first WebSocket frame received from a connecting client.
  • Trigger conditions: No authentication, session validation, or payload-type/schema checking is performed before deserialization — any client that can open a WebSocket connection to the /visual_register endpoint can trigger the vulnerable code path.
  • Attack vector: Network — an attacker only needs network reachability to the Config Server’s WebSocket listener; no credentials or user interaction are required.
  • Impact: Successful exploitation results in arbitrary code execution in the context of the Config Server process. Depending on deployment, this can lead to full host compromise, lateral movement into the broader LLM serving infrastructure, and exposure/tampering of any data or models accessible to that process.
  • Weakness class: CWE-502 (Deserialization of Untrusted Data).

Affected software

  • Product: LightLLM (ModelTC/lightllm)
  • Component: Config Server (/visual_register WebSocket endpoint)
  • Versions: All versions through 1.2.0 (inclusive) that expose the vulnerable visual_websocket_endpoint handler.

Severity

  • CVSS v3.1 Base Score: 9.8 (Critical) — CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
  • CVSS v4.0 Base Score: 9.3 (Critical) — CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Mitigation and recommended actions

  • Immediate: Upgrade LightLLM to a version that removes or fixes the unsafe deserialization in the Config Server’s /visual_register endpoint. Check the ModelTC/lightllm repository and release notes for a version beyond 1.2.0 that addresses this issue, and apply the update as soon as it is available.
  • If no patch is available:
    • Restrict network access to the Config Server’s WebSocket listener to trusted internal hosts only (e.g., via firewall rules, network segmentation, or VPN); do not expose it to the public internet.
    • Disable or block the /visual_register endpoint if the visual-service-registration feature is not in use.
    • Front the Config Server with an authenticating reverse proxy or gateway that validates client identity before allowing WebSocket connections to reach the application.
    • Monitor for unexpected or malformed WebSocket connection attempts to the Config Server as a potential indicator of exploitation attempts.

References

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

  • Mapping of all Assets with this Technology
  • Identification of potentially exposed assets to this CVE
  • Confirmation of verified exploitable assets
Get Exposure Report

How IONIX’s External Exposure Management Platform Detects and Validates
Zero-Days to Shrink MTTR

1

Map your entire attack surface (continously)

IONIX uses multi-factor discovery methods, including DNS analysis, certificate mapping, metadata inspection, and more, to automatically map every internet-facing asset across your environment. This includes cloud instances, third-party platforms, shadow IT, and even forgotten infrastructure that traditional tools miss.

2

Monitor for new CVEs

Dozens of threat intel feeds using agentic technology are continuously analyzed to detect the appearance of proof-of-concept code, exploit kits, and indicators of active targeting. IONIX goes further by applying AI to proactively evaluate whether emerging vulnerabilities are likely to be exploited, even before PoCs go public.

3

Identify Potential External Exposures

Not all CVEs matter. IONIX filters vulnerabilities by asking attacker-centric questions: Can it be reached from the internet? Does it require authentication? Is it being exploited in the wild? This dramatically reduces noise and focuses teams on threats that can actually be weaponized.

4

Create Safe, Scalable Exploit Validations

IONIX transforms real-world PoCs into safe, non-intrusive test payloads that can be run in production environments without disruption. These simulations are precisely targeted to the systems that are vulnerable, ensuring rapid validation without unnecessary load.

5

Execute Exploit Validations

By combining context about software stack, versioning, exposure status, and reachability, IONIX ensures that only the right payloads are executed against the right assets, maximizing efficiency and minimizing risk.

6

Drive Fast and Actionable Remediation

Results are routed through integrations with ticketing, SOAR, and SIEM tools. Issues are written in plain language, bundled into remediation clusters, and prioritized based on asset criticality, exploitability, and blast radius. This shortens mean time to remediation (MTTR) and empowers teams to act with confidence.

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

  • Mapping of all Assets with this Technology
  • Identification of potentially exposed assets to this CVE
  • Confirmation of verified exploitable assets
Get Exposure Report
Close

Subscribe to Threat Center RSS

Copy/paste the link below into your preferred RSS reader or follow these instructions to subscribe to Slack alerts.

Open RSS feedSlack setup instructions

Close

Get Real-Time CVE Alerts to Your Email

Be the first to know when new zero-days emerge