CVE-2026-34159

CRITICALNVD 9.89.8
EchelonGraph scoreMEDIUM confidence

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

This critical-severity CVE scores 9.8 under NVD CVSS v3. EPSS exploit probability: 1.6% (76th percentile of EPSS-scored CVEs). No GitHub Security Advisory CVSS is on record for this CVE.

Triggered by: Headline CVSS from NVD CVSS v3
Sources: EPSS, NVD CVSS v3
Elevated

Also known asEUVD-2026-17975 (ENISA EUVD)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).
  • Public exploit or proof-of-concept code is available
CISA-KEV: Not listedEPSS PROB: 1.6%CVSS: 9.8Exploit: Public PoCExposed services: Not assessed

A fix is available — apply it.

llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserialize_tensor() skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages. Combined with pointer leaks from ALLOC_BUFFER/BUFFER_GET_BASE, this gives full ASLR bypass and remote code execution. No authentication required, just TCP access to the RPC server port. This issue has been patched in version b8492.

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
1.6%
EPSS %ILE
76th
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.NVD CVSS v3 · CVSS v3.0 or v3.1
EG Likelihood
1.6% 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

April 1, 2026

Last Modified

October 26, 2026

Advisory Details (3)

Auto-updated Nov 3, 2026
🔬 Proof of concept available. Patch available. Sources: github_commit, github_pr, github.
GitHub advisoryMentions a patch (unverified)🟡 PoC Available

Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend · Advisory · ggml-org/llama.cpp · GitHub

https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw
GitHub pull request

RPC GRAPH_COMPUTE Arbitrary RCE patch

Fix merged in ggml-org/llama.cpp PR #20908 on 2026-03-23 — awaiting tagged release

https://github.com/ggml-org/llama.cpp/pull/20908
GitHub commit

commit 39bf0d3c6a95 (ggml-org/llama.cpp)

Fix landed in ggml-org/llama.cpp commit 39bf0d3c6a95 — awaiting tagged release

https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042

Affected Packages

(1 across 1 ecosystem)
Debian:14(1)
PackageVulnerable rangeFix by version rangeDependents
llama.cpp5151+dfsg-1~exp2 ... 8461+dfsg-1 (26 versions)
  • every version up to 8611+dfsg-1: fixed in 8611+dfsg-1
—

Weakness Classification(1)

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

Data Freshness Timeline

(refreshed 7× in last 7d / 30× 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 250 rows on record for this CVE, as 99 entries (1 repeat merged).

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. 1 repeat row among the latest 100 is merged into the row it repeats. The 150 oldest rows on record are not shown and were not checked, so the counts can still include repeats.

  1. 2026-11-12 16:33 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  2. 2026-11-12 09:38 UTCEchelonGraph re-scored
  3. 2026-11-11 16:39 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  4. 2026-11-11 07:14 UTCEchelonGraph re-scored
  5. 2026-11-10 16:43 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  6. 2026-11-10 04:48 UTCEchelonGraph re-scored
  7. 2026-11-09 16:35 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  8. 2026-11-09 02:25 UTCEchelonGraph re-scored
  9. 2026-11-08 16:39 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  10. 2026-11-08 00:02 UTCEchelonGraph re-scored
  11. 2026-11-07 16:38 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  12. 2026-11-06 21:38 UTCEchelonGraph re-scored
  13. 2026-11-06 16:37 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  14. 2026-11-06 06:25 UTCOSV refresh
  15. 2026-11-05 19:15 UTCEchelonGraph re-scored
  16. 2026-11-05 16:37 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  17. 2026-11-04 16:51 UTCEchelonGraph re-scored
  18. 2026-11-04 16:35 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  19. 2026-11-03 16:47 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  20. 2026-11-03 14:27 UTCEchelonGraph re-scored
  21. 2026-11-02 16:44 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  22. 2026-11-02 11:49 UTCEchelonGraph re-scored
  23. 2026-11-01 09:26 UTCEchelonGraph re-scored
  24. 2026-10-31 16:32 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  25. 2026-10-31 07:04 UTCEchelonGraph re-scored
Show 74 more
  1. 2026-10-30 16:34 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  2. 2026-10-30 04:39 UTCEchelonGraph re-scored
  3. 2026-10-29 16:33 UTCEPSS rescoreEPSS 1.6% · 76th percentile
  4. 2026-10-29 02:16 UTCEchelonGraph re-scored
  5. 2026-10-28 16:32 UTCEPSS rescoreEPSS 1.6% · 75th percentile
  6. 2026-10-27 23:51 UTCEchelonGraph re-scored
  7. 2026-10-27 16:31 UTCEPSS rescoreEPSS 1.6% · 75th percentile
  8. 2026-10-26 21:26 UTCEchelonGraph re-scored
  9. 2026-10-26 16:31 UTCEPSS rescoreEPSS 1.6% · 75th percentile
  10. 2026-10-25 18:07 UTCEchelonGraph re-scored
  11. 2026-10-25 16:32 UTCEPSS rescoreEPSS 1.6% · 75th percentile
  12. 2026-10-25 14:34 UTCEchelonGraph re-scored
  13. 2026-10-24 16:32 UTCEPSS rescoreEPSS 1.2% · 69th percentile
  14. 2026-10-24 15:23 UTCEchelonGraph re-scored
  15. 2026-10-23 16:31 UTCEPSS rescoreEPSS 1.2% · 69th percentile
  16. 2026-10-23 13:01 UTCEchelonGraph re-scored
  17. 2026-10-22 16:34 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  18. 2026-10-22 10:38 UTCEchelonGraph re-scored
  19. 2026-10-21 16:32 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  20. 2026-10-21 08:14 UTCEchelonGraph re-scored
  21. 2026-10-20 16:33 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  22. 2026-10-20 05:51 UTCEchelonGraph re-scored
  23. 2026-10-19 16:31 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  24. 2026-10-19 03:26 UTCEchelonGraph re-scored
  25. 2026-10-18 16:31 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  26. 2026-10-18 03:46 UTCEchelonGraph re-scored
  27. 2026-10-18 01:01 UTCEchelonGraph re-scored
  28. 2026-10-17 16:34 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  29. 2026-10-16 22:38 UTCEchelonGraph re-scored
  30. 2026-10-16 16:41 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  31. 2026-10-15 19:18 UTCEchelonGraph re-scored
  32. 2026-10-15 16:32 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  33. 2026-10-14 16:32 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  34. 2026-10-14 13:09 UTCEchelonGraph re-scored
  35. 2026-10-13 16:34 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  36. 2026-10-13 10:45 UTCEchelonGraph re-scored
  37. 2026-10-12 16:34 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  38. 2026-10-12 16:12 UTCEchelonGraph re-scored
  39. 2026-10-11 19:55 UTCEchelonGraph re-scored
  40. 2026-10-11 17:08 UTCEchelonGraph re-scored
  41. 2026-10-11 14:35 UTCEchelonGraph re-scored
  42. 2026-10-11 13:48 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  43. 2026-10-11 02:31 UTCOSV refresh
  44. 2026-10-10 20:00 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  45. 2026-10-09 20:07 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  46. 2026-10-08 17:12 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  47. 2026-10-07 23:47 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  48. 2026-10-06 15:17 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  49. 2026-10-05 17:42 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  50. 2026-10-04 23:22 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  51. 2026-10-03 14:25 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  52. 2026-10-01 19:49 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  53. 2026-09-30 15:03 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  54. 2026-09-28 13:51 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  55. 2026-09-27 13:48 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  56. 2026-09-26 15:58 UTCEPSS rescoreEPSS 1.2% · 68th percentile
  57. 2026-09-26 00:26 UTCEPSS rescoreEPSS 1.2% · 67th percentile
  58. 2026-09-21 14:19 UTCOSV refresh
  59. 2026-09-20 20:15 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  60. 2026-09-18 19:28 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  61. 2026-09-17 19:30 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  62. 2026-09-16 14:08 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  63. 2026-09-16 05:15 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  64. 2026-09-15 03:11 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  65. 2026-09-13 16:47 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  66. 2026-09-12 15:01 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  67. 2026-09-11 14:53 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  68. 2026-09-11 09:37 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  69. 2026-09-10 09:34 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  70. 2026-09-08 22:00 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  71. 2026-09-07 16:01 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  72. 2026-09-06 13:47 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  73. 2026-09-04 05:06 UTCEPSS rescoreEPSS 1.1% · 64th percentile
  74. 2026-09-02 14:12 UTCEPSS rescoreEPSS 1.1% · 64th percentile

Publicly available exploits

(1 reference)

Public exploit code is referenced for this CVE (1 GitHub PoC). Defenders should treat patch urgency accordingly.

  • GitHub PoCcasp3r0x0/CVE-2026-34159
    Published Apr 23, 2026

    0 Click RCE exploit for CVE-2026-34159 Lama.cpp RPC server

    Open source ↗

Frequently asked(5)

What is CVE-2026-34159?
CVE-2026-34159 is a critical-severity vulnerability; its CVE record was published on April 1, 2026. llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserialize_tensor() skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages.…
When was the CVE record for CVE-2026-34159 published?
The CVE record for CVE-2026-34159 was published on April 1, 2026. That is the record's publication date; the vulnerability itself may have been made public earlier. EchelonGraph's copy of the record carries a last-modified date of October 26, 2026; the record at its source may have been updated since.
Is CVE-2026-34159 actively exploited?
CVE-2026-34159 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 1.6% probability of exploitation in the next 30 days (76th percentile of EPSS-scored CVEs).
What is the CVSS score of CVE-2026-34159?
CVE-2026-34159 has a CVSS v3 base score of 9.8 (NVD).
How do I remediate CVE-2026-34159?
A fix for CVE-2026-34159 is available: update to the fixed version the vendor names in its advisory.

Dependency Blast Radius

See which npm, PyPI, Go, and Maven packages are affected by CVE-2026-34159

Explore →

Is Your Infrastructure Affected by CVE-2026-34159?

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