Skip to main content
VulnerabilityPublished

CVE-2026-124994: Turboderp-org ExLlamaV3 · Memory Buffer Overflow

Source GCVE Database

0
Compare
Authoritative record:CVE-2026-124994· Medium 5.3

Public exploit available · CVSS 5.3 Medium · fixed in 1.6.4 : patch ExLlamaV3 as a priority.

Data as of 10 Nov 2026, 01:35 UTC

CVSS 4.0 base score
5.3Medium
How severe the flaw is on the 0–10 CVSS scale.

Thin evidence

One scorer or none
CVSS scores by source
CNA (VulDB)CVSS 4.05.3
CNA (VulDB)CVSS 3.16.3
CNA (VulDB)CVSS 3.06.3
CNA (VulDB)CVSS 2.07.5
Patch
Available – fixed in 1.6.4
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 9 Nov 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponized: evidence found.

    CVSS exploit maturity: proof of concept public

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), CVSS vector, the Metasploit, Exploit-DB and nuclei-templates index as of 9 Nov 2026.

What to do

Affected software

ProductAffectedFixed in
Turboderp-orgExLlamaV3Affected
  • 1.6.0 – 1.6.3
4 affected builds
Fixed in–
GitHubturboderp-org/exllamav3pkg:github/turboderp-org/exllamav3Other threatsAffected
  • 1.6.0 – 1.6.3
4 affected builds
Fixed in–

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

Upgrade Turboderp-org ExLlamaV3 to version 1.6.4, which contains the official fix for this vulnerability. No other mitigation actions are indicated.

Description

A security flaw in Turboderp-org ExLlamaV3 up to version 1.6.3 involves an out-of-bounds read vulnerability in the ngram_dequant function of the NGramEmbedding component. The flaw is triggered by manipulation of the head_bias argument and can be exploited remotely. Public exploit code is available. Upgrading to version 1.6.4 addresses the issue.

References

Community reviews

Loading reviews…