CVE-2026-103270

waiting for further signalsZero needs more public information before it can validate it.

GHSA-x6wh-r9r9-mp7w ↗EUVD-2026-89989 ↗

no check availablenot proven exploitable
cve.orgpublic pocExploit code has been published by someone. Source signals can also carry a PoC (Proof of Concept).kevAttacks confirmed in the real world.
ModelTC LightLLM <= 1.2.0
CWE-306 Missing Authentication for Critical Function
ETK-2026-170, permanent identifier
20d 07h 23msince last run
56mAI analysisAnalysis: how long the AI spent reversing this signal.
timeline
0 = cve.org publication
trigger signal (cve.org) +0hwaiting for further signals+9h47m
0
+2h
+4h
+6h
+8h
+12h

description

LightLLM through version 1.2.0 may expose reinforcement-learning control functions without authentication. An unauthenticated remote party could interrupt inference activity or cause workers to become unresponsive when the service is started with reinforcement learning enabled.

mitigation

Disable reinforcement learning where it is not required, and do not expose the LightLLM HTTP API publicly. Restrict access to trusted network clients until a vendor-fixed release is available.

check
no
exploit
no
automationDeterministic: a check that runs on its own, the same request on every target. Contextual: an agentic technique that needs to be applied per target, for bugs that depend on contextual flows.
–
auth requiredWhether an attacker needs an account first, based on the CVSS vector's Privileges Required.
no
EPSSEstimate of the chance this CVE is exploited in the next 30 days. Refreshed daily.
0.51%42nd percentile
agent run time
56m
cve.org record, t = 0The moment cve.org publishes the CVE in full.
2026-09-30 14:48Z
Δ to public pocExploit code has been published by someone. Source signals can also carry a PoC (Proof of Concept).
–
Δ to kevAttacks confirmed in the real world.
–

outcome

No artifact: an exploit or check was attempted but could not be produced at this moment.

source signals

Public signals that triggered or updated the pipeline. These are inputs for the deltas above.

  1. 2026-09-30 14:48Z
    cve_org: trigger signal, started the run

this entry in the feed

GET /api/v1/feed/CVE-2026-103270

{
  "etk_id": "ETK-2026-170",
  "cve_id": "CVE-2026-103270",
  "display_id": "CVE-2026-103270",
  "ghsa_id": "GHSA-x6wh-r9r9-mp7w",
  "euvd_id": "EUVD-2026-89989",
  "status": "published",
  "anchor_state": "anchored",
  "title": "Missing authentication in LightLLM reinforcement-learning controls",
  "summary": "LightLLM through version 1.2.0 may expose reinforcement-learning control functions without authentication. An unauthenticated remote party could interrupt inference activity or cause workers to become unresponsive when the service is started with reinforcement learning enabled.",
  "mitigation": "Disable reinforcement learning where it is not required, and do not expose the LightLLM HTTP API publicly. Restrict access to trusted network clients until a vendor-fixed release is available.",
  "vendor": "ModelTC",
  "product": "LightLLM",
  "affected": [
    {
      "affected_notes": [],
      "affected_ranges": [
        ">= 0, <= 1.2.0"
      ],
      "ecosystem": null,
      "fixed_versions": [],
      "package_name": "LightLLM",
      "purl": null
    }
  ],
  "cwes": [
    306
  ],
  "cvss_score": 8.7,
  "cvss_src": "cna_vendor",
  "cvss_vector": "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",
  "epss_score": 0.0051,
  "epss_percentile": 0.4257,
  "epss_asof": "2026-10-20",
  "automation": "deterministic",
  "check": false,
  "check_confidence_basis": null,
  "exploit": false,
  "pipeline_minutes": 56.2075048,
  "t0_at": "2026-09-30T14:48:55.071000+00:00",
  "signals": {
    "trigger_signal": {
      "at": "2026-09-30T14:48:55.071000+00:00",
      "days": 0,
      "is_lower_bound": false,
      "measured_at": "2026-09-30T14:48:55.071000+00:00",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-103270"
    }
  }
}

published 2026-09-30 23:39Z
last change 2026-10-21 04:45Z

Need the technical details behind an entry?

Every entry here shows what we attempted and when. The technical details, the generated exploit that fired it and the evidence behind the verdict are not public. Tell us about your use case and we’ll get back to you about extended API access.