Skip to main content
VulnerabilityPublished Updated

CVE-2026-93348: unslothai unsloth · Code Injection

Affects unslothai unslothSource CVE Database V5

0
Compare

Not known to be exploited · CVSS 8.6 High : upgrade to 2026.8.20 soon.

Data as of 12 Nov 2026, 23:25 UTC

CVSS 4.0 base score
8.6High
How severe the flaw is on the 0–10 CVSS scale.
EPSS
0.4% chance of exploitation in 30 days – higher than 35% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.
Patch
Available – fixed in 2026.8.14, 2026.8.20
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 28 Sep 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponized: no evidence on record.

    No public exploit code indexed in Metasploit or Exploit-DB

  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), the Metasploit, Exploit-DB and nuclei-templates index as of 13 Nov 2026.

What to do

Upgrade unsloth to 2026.8.20; unsloth-zoo to 2026.8.14.

Affected software

ProductAffectedFixed in
unslothaiunsloth-zooAffected
  • >=2025.9.9 <2026.8.14
Fixed in
  • 2026.8.14
unslothaiunslothAffected
  • >=2025.9.9 <2026.8.20
Fixed in
  • 2026.8.20
GitHubunsloth-zoopkg:github/unsloth-zooOther threatsAffected
  • >=2025.9.9 <2026.8.14
Fixed in
  • 2026.8.14
GitHubunslothpkg:github/unslothOther threatsAffected
  • >=2025.9.9 <2026.8.20
Fixed in
  • 2026.8.20

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

AI-suggested

If you cannot patch yet, avoid loading untrusted or unauthenticated models in Unsloth Zoo.

Description

Unsloth Zoo versions from 2025.9.9 up to but not including 2026.8.14 contain a code injection vulnerability in the model-loading compile path. The vulnerability arises because the get_transformers_model_type() function does not enforce a character allowlist on model_type values, allowing malicious input to inject arbitrary Python code. This can lead to remote code execution when a malicious model is loaded for training or inference.

References

Community reviews

Loading reviews…