CVE-2026-103395

CRITICALPre-NVD 9.89.8—
EchelonGraph scoreHIGH confidence

Score 9.8 from GitHub Security Advisory (severity: CRITICAL) published 2026-09-30. A secondary CVSS source baseline 9.8; sources differ by 0.0.

Triggered by: GitHub Security Advisory CVSS
Sources: epss, ghsa, secondary
Trending — 4 sources updated this weekElevated
9.8EG
EchelonGraph verdictPlan mitigationSerious severity, but no confirmed exploitation yet.
  • High severity, but no confirmed exploitation yet
CISA-KEV: Not listedEPSS PROB: 0.6%CVSS: 9.8Exploit: Elevated riskExposed services: Not assessed

No fix is confirmed yet — apply a workaround or compensating control (WAF / firewall / segmentation) and watch for the fix.

LightLLM through 1.2.0 visual_only deployments expose an unauthenticated RPyC service with allow_pickle enabled that deserializes attacker-supplied arguments in the remote_infer_images method. Attackers can reach the visual RPyC port and pass objects with __reduce__ methods to execute arbitrary code with service account privileges.

CVSS v3
9.8
EG Score
9.8CRITICALhigh confidence
EG Risk
75
EG Risk 75/100CISA SSVC

EG Risk is EchelonGraph's 0–100 priority score: it fuses intrinsic severity with real-world exploitation and automatability so you can rank equal-severity CVEs and fix the most dangerous first. Higher = act sooner. Distinct from the 0–10 EG Score (severity).

How it’s computed
Severity98% × 45%
Exploitation40% × 40%
Automatability100% × 15%
CISA SSVC: Track at low, Track* at medium and Attend at high mission impact.
Action: No fix is confirmed yet. Restrict network exposure of the affected system or apply the vendor's mitigation within your standard update timelines at low or medium mission impact and sooner than that at high, and watch the vendor's advisory for the fix.
EPSS PROB
0.6%
EPSS %ILE
47th
KEV
Not listed

CISA SSVCTrack at low, Track* at medium and Attend at high mission impact.

No fix is confirmed yet. Restrict network exposure of the affected system or apply the vendor's mitigation within your standard update timelines at low or medium mission impact and sooner than that at high, and watch the vendor's advisory for the fix.

Exploitation public PoC (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact total (CISA Vulnrichment). Mission impact is CISA's Mission & Well-being decision point, and only you can judge it: high means the affected system is essential to your organisation's mission, or its compromise could cause irreversible harm to people. CISA's decision table

Published

September 30, 2026

Last Modified

September 30, 2026

Advisory Details (5)

Auto-updated Sep 30, 2026
No patch confirmed yet.
generic

[BUG] Unauthenticated RCE via pickle deserialization in the visual_only RPyC service · Issue #1610 · ModelTC/LightLLM · GitHub

https://github.com/ModelTC/LightLLM/issues/1610
generic

LightLLM through 1.2.0 Unauthenticated Remote Code Execution via Visual-Only RPyC Service | Advisories | VulnCheck

https://www.vulncheck.com/advisories/lightllm-through-1.2.0-unauthenticated-remote-code-execution-via-visual-only-rpyc-service
generic

LightLLM/lightllm/server/visualserver/visual_only_manager.py at v1.2.0 · ModelTC/LightLLM · GitHub

https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/visualserver/visual_only_manager.py#L138-L140
generic

LightLLM/lightllm/server/visualserver/objs.py at v1.2.0 · ModelTC/LightLLM · GitHub

https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/visualserver/objs.py#L6-L11
generic

GitHub - ModelTC/LightLLM: LightLLM is a Python-based LLM (Large Language Model) inference and serving framework, notable for its lightweight design, easy scalability, and high-speed performance. · GitHub

https://github.com/ModelTC/LightLLM

Vendor Advisories for CVE-2026-103395(1)

These vendors published their own advisory mentioning this CVE — often with vendor-specific remediation steps + affected product lists not in NVD.

Weakness Classification(1)

MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.

Data Freshness Timeline

(refreshed 38× in last 7d / 38× in last 30d)

Each row is a source pipeline that fetched or updated this CVE on that date, with what changed. For example, "NVD update" means NVD published or revised its analysis for this CVE; "MITRE cvelistV5" means we ingested or refreshed it from the CNA feed. Most recent first.

  1. 2026-10-04 09:00 UTCGHSA enrichment
  2. 2026-10-04 05:20 UTCGHSA enrichment
  3. 2026-10-04 01:40 UTCGHSA enrichment
  4. 2026-10-03 21:59 UTCGHSA enrichment
  5. 2026-10-03 18:19 UTCGHSA enrichment
  6. 2026-10-03 14:37 UTCEG score recompute
  7. 2026-10-03 14:36 UTCGHSA enrichment
  8. 2026-10-03 14:25 UTCEPSS rescore
  9. 2026-10-03 10:56 UTCGHSA enrichment
  10. 2026-10-03 07:17 UTCGHSA enrichment
  11. 2026-10-03 03:37 UTCGHSA enrichment
  12. 2026-10-02 23:57 UTCGHSA enrichment
  13. 2026-10-02 20:17 UTCGHSA enrichment
  14. 2026-10-02 16:34 UTCEG score recompute
  15. 2026-10-02 16:34 UTCGHSA enrichment
  16. 2026-10-02 12:54 UTCGHSA enrichment
  17. 2026-10-02 09:15 UTCGHSA enrichment
  18. 2026-10-02 05:35 UTCGHSA enrichment
  19. 2026-10-02 01:52 UTCGHSA enrichment
  20. 2026-10-01 22:12 UTCEG score recompute
  21. 2026-10-01 22:12 UTCGHSA enrichment
  22. 2026-10-01 19:49 UTCEPSS rescore
  23. 2026-10-01 18:31 UTCGHSA enrichment
  24. 2026-10-01 14:51 UTCGHSA enrichment
  25. 2026-10-01 11:11 UTCGHSA enrichment
Show 13 more
  1. 2026-10-01 07:32 UTCGHSA enrichment
  2. 2026-10-01 03:53 UTCGHSA enrichment
  3. 2026-10-01 00:14 UTCEG score recompute
  4. 2026-10-01 00:14 UTCGHSA enrichment
  5. 2026-09-30 20:22 UTCEG score recompute
  6. 2026-09-30 20:22 UTCGHSA enrichment
  7. 2026-09-30 19:46 UTCEG score recompute
  8. 2026-09-30 19:46 UTCGHSA enrichment
  9. 2026-09-30 19:16 UTCGHSA enrichment
  10. 2026-09-30 15:37 UTCEG score recompute
  11. 2026-09-30 15:37 UTCGHSA enrichment
  12. 2026-09-30 14:52 UTCEG score recompute
  13. 2026-09-30 14:52 UTCMITRE cvelistV5first tracked

Related CVEs(same CWE)

Frequently asked(5)

What is CVE-2026-103395?
CVE-2026-103395 is a critical vulnerability published on September 30, 2026. LightLLM through 1.2.0 visualonly deployments expose an unauthenticated RPyC service with allowpickle enabled that deserializes attacker-supplied arguments in the remoteinferimages method. Attackers can reach the visual RPyC port and pass objects with reduce methods to execute arbitrary code with…
When was CVE-2026-103395 disclosed?
CVE-2026-103395 was first published on September 30, 2026. EchelonGraph re-ingests CVE updates from NVD on a 2-hour cycle, so this page reflects the latest published state.
Is CVE-2026-103395 actively exploited?
CVE-2026-103395 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 0.6% probability of exploitation in the next 30 days (47th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-103395?
CVE-2026-103395 has a CVSS base score of 9.8 (a secondary CVSS source that NVD displays; NVD's own analysis pending).
How do I remediate CVE-2026-103395?
No fix for CVE-2026-103395 is confirmed yet. Until one is published, restrict network exposure of the affected system or apply the vendor's mitigation — for example, keep it off the internet or limit it to trusted networks — and watch the vendor's advisory for the fix. The vendor advisories EchelonGraph has for CVE-2026-103395 are linked in the Vendor Advisories panel on this page.

Dependency Blast Radius

Explore the affected products and dependency analysis for CVE-2026-103395

Explore →

Is Your Infrastructure Affected by CVE-2026-103395?

EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.

Book a Demo →Browse All CVEs