CVE-2026-119052

validatingZero is analyzing and reversing it now.

GHSA-rr32-94xq-hrxw ↗EUVD-2026-112818 ↗

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.
ggml-org llama.cpp >= b2877, < b11935
fixed in b11935
CWE-248 Uncaught Exception
ETK-2026-2972Zero's internal tracking ID, independent of the CVE number., permanent identifier
–running
startingcurrent stage
timeline
0 = cve.org publication
trigger signal (cve.org) +0h
0
+2h
+4h
+6h
+8h
+10h
+12h

description

llama.cpp ggml-rpc servers from b2877 through releases before b11935 can be terminated by malformed oversized request length data. An unauthenticated remote attacker can deny service to connected clients if they can reach the rpc-server and establish a protocol session.

Still under analysis: more detail becomes available as the pipeline runs.

mitigation

Upgrade llama.cpp to b11935 or later. Restrict network access to the ggml-rpc server to trusted clients until the upgrade is complete.

fixed in b11935

check
building now
exploit
not yet
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.
–
agent run time
–
cve.org record, t = 0The moment cve.org publishes the CVE in full.
2026-10-29 13:08Z
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

Run in progress: no artifact yet.

source signals

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

  1. 2026-10-29 13:08Z
    cve_org: trigger signal, started the run

this entry in the feed

GET /api/v1/feed/CVE-2026-119052

{
  "etk_id": "ETK-2026-2972",
  "cve_id": "CVE-2026-119052",
  "display_id": "CVE-2026-119052",
  "ghsa_id": "GHSA-rr32-94xq-hrxw",
  "euvd_id": "EUVD-2026-112818",
  "status": "in_pipeline",
  "anchor_state": "anchored",
  "title": "Unauthenticated denial of service in llama.cpp ggml-rpc server",
  "summary": "llama.cpp ggml-rpc servers from b2877 through releases before b11935 can be terminated by malformed oversized request length data. An unauthenticated remote attacker can deny service to connected clients if they can reach the rpc-server and establish a protocol session.",
  "mitigation": "Upgrade llama.cpp to b11935 or later. Restrict network access to the ggml-rpc server to trusted clients until the upgrade is complete.",
  "vendor": "ggml-org",
  "product": "llama.cpp",
  "affected": [
    {
      "affected_notes": [],
      "affected_ranges": [
        ">= b2877, < b11935"
      ],
      "ecosystem": null,
      "fixed_versions": [
        "b11935"
      ],
      "package_name": "llama.cpp",
      "purl": null
    }
  ],
  "cwes": [
    248
  ],
  "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": null,
  "epss_percentile": null,
  "epss_asof": null,
  "automation": null,
  "check": false,
  "check_confidence_basis": null,
  "exploit": false,
  "pipeline_minutes": null,
  "t0_at": "2026-10-29T13:08:52.886000+00:00",
  "signals": {
    "trigger_signal": {
      "at": "2026-10-29T13:08:52.886000+00:00",
      "days": 0,
      "is_lower_bound": false,
      "measured_at": "2026-10-29T13:08:52.886000+00:00",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-119052"
    }
  }
}

published 2026-10-29 13:52Z
last change 2026-10-29 15:55Z

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.