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

Back
New CVE Detected

CVE-2026-103270 – Unauthenticated Denial of Service via Exposed RL Control Endpoints – LightLLM ≤ 1.

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

Subscribe to Email Alerts
Subscribe to RSS Feed

Summary

CVE-2026-103270 is a missing authentication vulnerability (CWE-306) in ModelTC’s LightLLM, an open-source LLM inference serving framework. LightLLM versions through 1.2.0 mount a set of reinforcement-learning (RL) control routes on the public HTTP API without any credential check, letting an unauthenticated, remote attacker disrupt an active inference cluster. The flaw carries a High severity CVSS v3.1 score of 7.5 (CVSS v4.0: 8.7), reflecting a purely availability-impacting but easily exploited network attack.

Technical details

  • Root cause: lightllm/server/api_http.py unconditionally imports and registers the RL router (api_http_rl.py) on the main FastAPI application. None of the handlers in that router validate any form of credential, and LightLLM has no authentication layer at all.
  • Affected endpoints: /pause_generation, /continue_generation, /abort_request, /flush_cache, /release_memory_occupation, /resume_memory_occupation, /init_weights_update_group, /destroy_weights_update_group, /update_weights_from_distributed, /update_weights_from_tensor, and /update_weights_from_ipc.
  • Trigger conditions: The routes are mounted regardless of whether the server was started with the --enable_rl flag. When RL mode is enabled, calls to these endpoints are processed and take effect immediately (pausing generation, aborting requests, flushing the KV cache, or joining an attacker-chosen NCCL process group). When RL mode is disabled, the same routes still respond (with HTTP 500/417 errors instead of 404), confirming they remain exposed and probeable from the network.
  • Attack vector: Network — any client able to reach the LightLLM HTTP port can send unauthenticated requests to these endpoints; no user interaction or prior access is required.
  • Impact: Denial of service against the inference cluster — attackers can halt all generation cluster-wide, terminate arbitrary in-flight requests, purge the prompt cache, release/occupy GPU memory, or wedge worker processes by forcing them into an attacker-controlled NCCL process group, requiring a restart to recover.

Affected software

  • ModelTC LightLLM — versions 0 through 1.2.0 (all versions up to and including 1.2.0)
  • Confirmed present on commit 1eb4810c (v1.2.0+50) and on the main branch as of the disclosure date

Severity

  • CVSS v3.1 Base Score: 7.5 (HIGH) — CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
  • CVSS v4.0 Base Score: 8.7 (HIGH) — CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N

Mitigation and recommended actions

  • Immediate: No official patched release has been published by ModelTC at the time of writing. Track the upstream GitHub issue (#1609) for a fix and upgrade to a patched version as soon as one is released.
  • Workarounds (no patch available):
    • Restrict network access to the LightLLM HTTP port using firewall rules or network segmentation so it is reachable only from trusted internal hosts/services.
    • Do not expose LightLLM instances directly to the internet or untrusted networks, especially deployments started with --enable_rl.
    • Place an authenticating reverse proxy or API gateway in front of the LightLLM service to block unauthenticated requests to /pause_generation, /abort_request, /flush_cache, /init_weights_update_group, and the other RL control routes listed above.
    • Monitor for unexpected calls to these endpoints as a possible 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