LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions prior to 0.12.3 have a Server-Side Request Forgery (SSRF) vulnerability in LMDeploy's vision-language module. The load_image() function in lmdeploy/vl/utils.py fetches arbitrary URLs without validating internal/private IP addresses, allowing attackers to access cloud metadata services, internal networks, and sensitive resources. Version 0.12.3 patches the issue.
CVE-2026-33626
“—” beside the score: it is equal to NVD's CVSS score.
Score 7.5 from GitHub Security Advisory (severity: HIGH) published 2026-04-21. NVD baseline CVSS 7.5; sources differ by 0.0.
- CISA's Vulnrichment rates exploitation as PoC; EchelonGraph holds no exploit-code reference for it
A fix is available — apply it.
- CVSS v3
- 7.5
- EG Score
- 7.5HIGHmedium confidence
- EG Risk
- 65EG Risk 65/100
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 (value × EchelonGraph’s weight)Severity75% × 45%The EG Score, 7.5 of 10.Exploitation40% × 40%EchelonGraph's fixed weight for EG-KEV Tier 3 (Elevated): a constant, not a probability.Automatability100% × 15%CISA's SSVC Automatable value: yes (Vulnrichment).SSVC (CISA's decision table, applied by EchelonGraph): Track at low or medium mission impact; Attend at high (mission-essential systems).Exploitation public PoC (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE.Action: A fix is available. Apply it within your standard update timelines at low or medium mission impact and sooner than that at high. - EPSS PROB
- 1.5%
- EPSS %ILE
- 74th
- KEV
- Not listed
SSVC (CISA's decision table, applied by EchelonGraph)Track at low or medium mission impact; Attend at high (mission-essential systems).
A fix is available. Apply it within your standard update timelines at low or medium mission impact and sooner than that at high.
Exploitation public PoC (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE. Mission impact is CISA's Mission & Well-being decision point, and only you can judge it. CISA's table makes it high when Mission Prevalence is Essential — the vulnerable component "directly provides capabilities that constitute at least one MEF for at least one entity" (MEF: mission essential function) — or when Public Well-Being Impact is Irreversible: "multiple fatalities are likely", the cyber-physical system "is likely lost or destroyed", "extreme or serious externalities" are imposed on other parties, or social systems such as elections or the financial grid "are destabilized and potentially collapse". CISA's decision table
BOD 26-04 (CISA's remediation timeline, Table 1)14 days if the affected system is publicly exposed; 60 days if it is not.
In CISA KEV no (CISA's KEV catalog) · Automatable yes (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). Whether your system is publicly exposed is yours to answer. BOD 26-04 binds US Federal Civilian Executive Branch agencies; for anyone else it is CISA's published timeline, a reference rather than an obligation. The directive · how EchelonGraph applies it
Published
April 20, 2026
Last Modified
April 23, 2026
Advisory Details (4)
Auto-updated Jun 9, 2026Server-Side Request Forgery (SSRF) in Vision-Language Image Loading · Advisory · InternLM/lmdeploy · GitHub
https://github.com/InternLM/lmdeploy/security/advisories/GHSA-6w67-hwm5-92mqv0.12.3
Patch available: InternLM/lmdeploy v0.12.3
https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3fix security issues
Fix merged in InternLM/lmdeploy PR #4447 on 2026-03-27 — awaiting tagged release
https://github.com/InternLM/lmdeploy/pull/4447commit 71d64a339edb (InternLM/lmdeploy)
Fix landed in InternLM/lmdeploy commit 71d64a339edb — awaiting tagged release
https://github.com/InternLM/lmdeploy/commit/71d64a339edb901e9005358e0633fbbab367d626Vendor Advisories for CVE-2026-33626(1)
These vendors published their own advisory mentioning this CVE — often with vendor-specific remediation steps + affected product lists not in NVD.
Patch Availability(1)
| Vendor / Ecosystem | Fix (by version range) / Patch | Released | Source |
|---|---|---|---|
| pip | lmdeploy | — | ghsa |
Patches are aggregated from vendor advisories (Red Hat, Microsoft, Cisco, GitHub) and package ecosystems (OSV, GHSA). Multiple rows for the same upstream release have been deduplicated.
Affected Packages
(1 across 1 ecosystem)
PyPI(1)
| Package | Vulnerable range | Fix by version range | Dependents |
|---|---|---|---|
| lmdeploy | 0.0.10 ... 0.9.2.post1 (51 versions) |
| — |
Weakness Classification(1)
MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.
Data Freshness Timeline
(refreshed 12× in last 7d / 28× in last 30d)
Each row is a time one of our pipelines fetched, updated or re-scored this CVE. 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. An arrow marks a change the record states ("CVSS v3 → 9.1"); EPSS and GHSA rows show the values recorded at that time, not what changed (EPSS values are rounded; hover one for the exact figures). Most recent first.
Showing the most recent 93 of 192 refreshes on record for this CVE.
A repeat, a row identical to one the same pipeline wrote less than 10 seconds earlier, is not a separate refresh and is not counted. 7 repeat rows among the latest 100 are merged into the row they repeat. The 99 oldest rows on record are not shown and were not checked, so the counts can still include repeats.
- 2026-10-16 16:41 UTCEPSS rescoreEPSS 1.5% · 74th percentile
- 2026-10-15 16:32 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-14 16:32 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-13 16:34 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-12 16:34 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-12 16:13 UTCEG score recompute
- 2026-10-11 19:55 UTCEG score recompute
- 2026-10-11 17:08 UTCEG score recompute
- 2026-10-11 14:36 UTCEG score recompute▼ 0.40
- 2026-10-11 13:48 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-10 20:00 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-10 18:33 UTCOSV refresh
- 2026-10-09 20:07 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-08 17:12 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-07 23:47 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-06 15:17 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-05 17:42 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-03 14:25 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-10-01 19:49 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-09-30 15:03 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-09-28 13:51 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-09-27 13:48 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-09-26 15:58 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-09-26 00:26 UTCEPSS rescoreEPSS 1.5% · 73rd percentile
- 2026-09-21 05:49 UTCOSV refresh
Show 68 moreShow fewer
- 2026-09-20 20:15 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-18 19:28 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-17 19:30 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-16 14:08 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-16 05:15 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-15 03:11 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-13 16:47 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-12 15:01 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-11 09:37 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-10 09:34 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-08 22:00 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-06 13:47 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-04 05:06 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-02 14:11 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-02 03:39 UTCOSV refresh
- 2026-09-01 13:53 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-09-01 04:39 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-30 19:17 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-30 01:22 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-28 21:41 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-27 14:25 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-26 14:46 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-24 14:17 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-21 23:49 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-20 22:55 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-19 17:03 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-18 13:48 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-17 13:47 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-16 06:22 UTCEG score recompute
- 2026-08-16 06:22 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-16 02:14 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-15 01:30 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-14 15:48 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-14 04:32 UTCEG score recompute
- 2026-08-14 04:32 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-13 22:00 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-13 08:48 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-12 18:43 UTCEG score recompute
- 2026-08-12 18:43 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-12 13:50 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-11 16:53 UTCEG score recompute
- 2026-08-11 16:53 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-11 02:29 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-10 02:16 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-09 15:00 UTCEG score recompute
- 2026-08-09 15:00 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-09 13:46 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-09 03:45 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-08 11:44 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-08 00:25 UTCEG score recompute
- 2026-08-08 00:25 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-07 13:09 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-06 21:49 UTCEG score recompute
- 2026-08-06 21:49 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-06 13:46 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-08-06 10:29 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-05 23:13 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-03 21:25 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-03 00:44 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-02 12:54 UTCEG score recompute
- 2026-08-02 12:54 UTCGHSA enrichmentGHSA-6w67-hwm5-92mq · GitHub severity HIGH · GitHub CVSS 7.5
- 2026-08-01 04:16 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-07-30 16:28 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-07-30 01:30 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-07-26 14:54 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-07-25 14:18 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-07-24 14:17 UTCEPSS rescoreEPSS 45% · 98th percentile
- 2026-07-23 14:18 UTCEPSS rescoreEPSS 45% · 98th percentile
Public detection templates
(1 reference)A public Nuclei detection template is referenced for this CVE. A detection template checks whether a system is affected; it is not necessarily exploit code.
- Nucleihttp/cves/2026/CVE-2026-33626.yamlPublished Apr 30, 2026
LMDeploy - Server-Side Request Forgery
Open source ↗
Related CVEs(same product + same CWE)
Same product
8 shownPyPI:lmdeploy
Frequently asked(5)
What is CVE-2026-33626?
When was the CVE record for CVE-2026-33626 published?
Is CVE-2026-33626 actively exploited?
What is the CVSS score of CVE-2026-33626?
How do I remediate CVE-2026-33626?
Dependency Blast Radius
See which npm, PyPI, Go, and Maven packages are affected by CVE-2026-33626
Is Your Infrastructure Affected by CVE-2026-33626?
EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.