CVE-2026-76850

CRITICALPre-NVD 9.89.8
EchelonGraph scoreMEDIUM confidence

“—” beside the score: it is equal to the CVSS score beside it.

Score 9.8 from GitHub Security Advisory (severity: CRITICAL) published 2026-08-20. A secondary CVSS source baseline 9.8; sources differ by 0.0.

Triggered by: GitHub Security Advisory CVSS
Sources: EPSS, GitHub Security Advisory, a secondary CVSS source
Elevated

Also known asEUVD-2026-62972 (ENISA EUVD) · GHSA-q66m-3fcg-jjh7 (GitHub Security Advisories)Source: ENISA EUVD

9.8EG
EchelonGraph verdictPlan a fixRemediate within your standard update timelines at low or medium mission impact and sooner than that at high (the SSVC decision on this page).
  • CISA's Vulnrichment rates exploitation as PoC; EchelonGraph holds no exploit-code reference for it
CISA-KEV: Not listedEPSS PROB: 1.3%CVSS: 9.8Exploit: No code on recordExposed services: Not assessed

A fix is available — apply it.

LMDeploy deserializes disaggregated-serving peer messages with pickle. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py reads peer-to-peer cache-free requests with recv_pyobj(), which deserializes the received bytes with pickle.loads(), and the isinstance check against DistServeCacheFreeRequest runs only after deserialization has already completed. The peer that supplies those bytes is caller-controlled: p2p_connect passes remote_engine_endpoint_info.zmq_address from the request body to connect() on the ZMQ PULL socket, and the POST /distserve/p2p_initialize and /distserve/p2p_connect endpoints in lmdeploy/serve/openai/api_server.py apply no authentication unless the server is started with api_keys, which defaults to None. A remote attacker can direct an engine to pull from a ZMQ endpoint under their control and execute arbitrary code in the engine process. Deployments that do not enable disaggregated serving are not affected, because the receive loop is only started once the migration backend accepts the connection.

CVSS v3
9.8
EG Score
9.8CRITICALmedium confidence
EG Risk
75
EG Risk 75/100

EG Risk is EchelonGraph's 0–100 priority score. It fuses impact severity (EG Severity, the CVSS base score, never the KEV-floored or EPSS-lifted EG Score) with real-world exploitation and automatability, under weights EchelonGraph set by hand, so you can rank equal-severity CVEs and fix the most dangerous first. Higher = act sooner.

How it’s computed (value × EchelonGraph’s weight)
Severity98% × 45%
Impact severity (EG Severity): CVSS 9.8 of 10. Never the KEV-floored or EPSS-lifted EG Score.
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, Track* at medium and Attend at high mission impact.
Exploitation public PoC (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact total (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
1.3%
EPSS %ILE
70th
KEV
Not listed

What the EG score is made of (scorer v4.0: the three questions the 0–10 score mixes, published apart · methodology)

The EG score (9.8): severity, escalated by confirmed exploitation. This CVE's 9.8 is its impact severity (EG Severity), not escalated.

EG Severity
9.8 CRITICALCVSS-B, impact only: no KEV listing, EPSS or EG-KEV tier raises it.GitHub Security Advisory CVSS
EG Likelihood
1.3% within 30 daysSource: FIRST EPSS (model v2026.06.15, score date 2026-11-11). FIRST's exploitation probability, not an EchelonGraph model.
Confirmed exploitation
None on record (CISA KEV, EG-KEV Tier 1)Reported apart from the likelihood: an observed fact, not a probability.
EG Priority
SSVC Track* at medium mission impactTrack at low, Track* at medium and Attend at high mission impact.CISA BOD 26-04: 3 days if the affected system is publicly exposed; 60 days if it is not.Each input and its source, and whom BOD 26-04 binds: the SSVC and BOD 26-04 blocks below.

SSVC (CISA's decision table, applied by EchelonGraph)Track at low, Track* at medium and Attend at high mission impact.

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 total (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)3 days if the affected system is publicly exposed; 60 days if it is not.

The score's inputs: In CISA KEV no (CISA's KEV catalog) · Automatable yes (CISA Vulnrichment) · Technical impact total (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

August 19, 2026

Last Modified

September 24, 2026

Advisory Details (7)

Auto-updated Nov 8, 2026
Patch available. Sources: github_commit, github_release.
Web page

[Bug] Unauthenticated pickle deserialization via ZMQ in disaggregated serving → RCE · Issue #4804 · InternLM/lmdeploy · GitHub

https://github.com/InternLM/lmdeploy/issues/4804
Web page

LMDeploy Remote Code Execution via Unsafe Pickle Deserialization in the Disaggregated Serving Peer Connector | Advisories | VulnCheck

https://www.vulncheck.com/advisories/lmdeploy-remote-code-execution-via-unsafe-pickle-deserialization-in-the-disaggregated-serving-peer-connector
GitHub release Patch Available

v0.16.0

Patch available: InternLM/lmdeploy v0.16.0

https://github.com/InternLM/lmdeploy/releases/tag/v0.16.0
GitHub commit

commit f05b4ad8bf2e (InternLM/lmdeploy)

Fix landed in InternLM/lmdeploy commit f05b4ad8bf2e — awaiting tagged release

https://github.com/InternLM/lmdeploy/commit/f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2
Web page

lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py at v0.15.0 · InternLM/lmdeploy · GitHub

https://github.com/InternLM/lmdeploy/blob/v0.15.0/lmdeploy/pytorch/disagg/conn/engine_conn.py#L79
Web page

lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py at v0.15.0 · InternLM/lmdeploy · GitHub

https://github.com/InternLM/lmdeploy/blob/v0.15.0/lmdeploy/pytorch/disagg/conn/engine_conn.py#L61
Web page

GitHub - InternLM/lmdeploy: LMDeploy is a toolkit for compressing, deploying, and serving LLMs. · GitHub

https://github.com/InternLM/lmdeploy

Vendor Advisories for CVE-2026-76850(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 14× in last 7d / 59× 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 its CVE.org record (CNA or CISA ADP). 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). "EchelonGraph re-scored" is our own recompute of this CVE's score, not an update from a source, and is not counted in the refreshes above. Most recent first.

Showing the most recent 100 of 445 rows 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. None are among the latest 100 rows. The 345 oldest rows on record are not shown and were not checked, so the counts can still include repeats.

  1. 2026-11-12 16:34 UTCEPSS rescoreEPSS 1.3% · 70th percentile
  2. 2026-11-12 14:31 UTCEchelonGraph re-scored
  3. 2026-11-12 14:31 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  4. 2026-11-11 16:40 UTCEPSS rescoreEPSS 1.3% · 70th percentile
  5. 2026-11-11 13:27 UTCEchelonGraph re-scored
  6. 2026-11-11 13:27 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  7. 2026-11-10 16:45 UTCEPSS rescoreEPSS 1.3% · 70th percentile
  8. 2026-11-10 12:23 UTCEchelonGraph re-scored
  9. 2026-11-10 12:22 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  10. 2026-11-09 16:36 UTCEPSS rescoreEPSS 1.3% · 70th percentile
  11. 2026-11-09 11:18 UTCEchelonGraph re-scored
  12. 2026-11-09 11:18 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  13. 2026-11-08 16:40 UTCEPSS rescoreEPSS 1.3% · 70th percentile
  14. 2026-11-08 10:14 UTCEchelonGraph re-scored
  15. 2026-11-08 10:14 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  16. 2026-11-07 16:39 UTCEPSS rescoreEPSS 1.3% · 70th percentile
  17. 2026-11-07 09:06 UTCEchelonGraph re-scored
  18. 2026-11-07 09:06 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  19. 2026-11-06 18:25 UTCOSV refresh
  20. 2026-11-06 16:39 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  21. 2026-11-06 08:03 UTCEchelonGraph re-scored
  22. 2026-11-06 08:03 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  23. 2026-11-05 16:38 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  24. 2026-11-05 06:59 UTCEchelonGraph re-scored
  25. 2026-11-05 06:59 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
Show 75 more
  1. 2026-11-04 16:36 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  2. 2026-11-04 05:55 UTCEchelonGraph re-scored
  3. 2026-11-04 05:55 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  4. 2026-11-03 16:49 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  5. 2026-11-03 04:51 UTCEchelonGraph re-scored
  6. 2026-11-03 04:51 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  7. 2026-11-02 16:46 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  8. 2026-11-02 03:46 UTCEchelonGraph re-scored
  9. 2026-11-02 03:46 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  10. 2026-11-01 16:35 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  11. 2026-11-01 02:43 UTCEchelonGraph re-scored
  12. 2026-11-01 02:43 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  13. 2026-10-31 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  14. 2026-10-31 01:39 UTCEchelonGraph re-scored
  15. 2026-10-31 01:39 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  16. 2026-10-30 16:35 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  17. 2026-10-30 00:35 UTCEchelonGraph re-scored
  18. 2026-10-30 00:35 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  19. 2026-10-29 16:34 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  20. 2026-10-28 23:32 UTCEchelonGraph re-scored
  21. 2026-10-28 23:32 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  22. 2026-10-28 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  23. 2026-10-27 22:29 UTCEchelonGraph re-scored
  24. 2026-10-27 22:28 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  25. 2026-10-27 16:32 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  26. 2026-10-26 21:21 UTCEchelonGraph re-scored
  27. 2026-10-26 21:21 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  28. 2026-10-26 16:32 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  29. 2026-10-25 18:13 UTCEchelonGraph re-scored
  30. 2026-10-25 18:13 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  31. 2026-10-25 16:34 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  32. 2026-10-25 14:01 UTCEchelonGraph re-scored
  33. 2026-10-24 16:59 UTCEchelonGraph re-scored
  34. 2026-10-24 16:59 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  35. 2026-10-24 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  36. 2026-10-23 16:32 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  37. 2026-10-23 15:55 UTCEchelonGraph re-scored
  38. 2026-10-23 15:55 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  39. 2026-10-22 16:35 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  40. 2026-10-22 14:49 UTCEchelonGraph re-scored
  41. 2026-10-22 14:48 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  42. 2026-10-21 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  43. 2026-10-21 13:44 UTCEchelonGraph re-scored
  44. 2026-10-21 13:44 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  45. 2026-10-20 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  46. 2026-10-20 12:40 UTCEchelonGraph re-scored
  47. 2026-10-20 12:40 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  48. 2026-10-19 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  49. 2026-10-19 11:35 UTCEchelonGraph re-scored
  50. 2026-10-19 11:35 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  51. 2026-10-18 16:32 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  52. 2026-10-18 09:22 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  53. 2026-10-18 03:12 UTCEchelonGraph re-scored
  54. 2026-10-17 16:35 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  55. 2026-10-17 08:17 UTCEchelonGraph re-scored
  56. 2026-10-17 08:17 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  57. 2026-10-16 16:44 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  58. 2026-10-16 07:12 UTCEchelonGraph re-scored
  59. 2026-10-16 07:12 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  60. 2026-10-15 16:32 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  61. 2026-10-15 06:01 UTCEchelonGraph re-scored
  62. 2026-10-15 06:01 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  63. 2026-10-14 16:33 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  64. 2026-10-14 01:40 UTCEchelonGraph re-scored
  65. 2026-10-14 01:39 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  66. 2026-10-13 16:35 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  67. 2026-10-12 21:50 UTCEchelonGraph re-scored
  68. 2026-10-12 21:50 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  69. 2026-10-12 16:35 UTCEPSS rescoreEPSS 1.3% · 69th percentile
  70. 2026-10-12 15:38 UTCEchelonGraph re-scored
  71. 2026-10-12 05:30 UTCGHSA enrichmentGHSA-q66m-3fcg-jjh7 · GitHub severity CRITICAL · GitHub CVSS 9.8
  72. 2026-10-11 19:23 UTCEchelonGraph re-scored
  73. 2026-10-11 16:39 UTCEchelonGraph re-scored
  74. 2026-10-11 13:58 UTCEchelonGraph re-scored
  75. 2026-10-11 13:49 UTCEPSS rescoreEPSS 1.3% · 69th percentile

Related CVEs(same CWE)

Frequently asked(5)

What is CVE-2026-76850?
CVE-2026-76850 is a critical-severity vulnerability; its CVE record was published on August 19, 2026. LMDeploy deserializes disaggregated-serving peer messages with pickle. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py reads peer-to-peer cache-free requests with recv_pyobj(), which deserializes the received bytes with pickle.loads(), and the isinstance check against…
When was CVE-2026-76850 made public?
CVE-2026-76850 was made public on July 29, 2026: that is the date its CVE record's CNA, VulnCheck, gives for when the vulnerability was made public. The CVE record itself was published on August 19, 2026. EchelonGraph's copy of the record carries a last-modified date of September 24, 2026; the record at its source may have been updated since.
Is CVE-2026-76850 actively exploited?
CVE-2026-76850 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 1.3% probability of exploitation in the next 30 days (70th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-76850?
CVE-2026-76850 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-2026-76850?
A fix for CVE-2026-76850 is available: update to the fixed version the vendor names in its advisory. The vendor advisories EchelonGraph has for CVE-2026-76850 are linked in the Vendor Advisories panel on this page.

Dependency Blast Radius

Explore the affected products and dependency analysis for CVE-2026-76850

Explore →

Is Your Infrastructure Affected by CVE-2026-76850?

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

Book a Demo →Browse All CVEs