CVE-2026-86169
Axolotl through 0.18.0 contains a remote code execution vulnerability in the multipack patch path where trust_remote_code defaults to None instead of False, causing the security guard to be bypassed. Attackers can execute arbitrary Python code by crafting a malicious Hugging Face model repository selected as base_model, which is loaded with hardcoded trust_remote_code=True during AutoModelForCausalLM.from_pretrained.
CWE-829 : Inclusion of Functionality from Untrusted Control Sphere
Default Status: unaffected
github.com: Pull Request #3858 external site patch issue-tracking
github.com: Patch Commit external site patch
github.com: Multipack trust_remote_code guard external site technical-description
github.com: patch_remote sink external site technical-description
external site product
vulncheck.com: VulnCheck Advisory: Axolotl through 0.18.0 Remote Code Execution via Multipack Patching external site third-party-advisory
The full story
This article is one source in a clustered incident — the cluster page carries the summary, timeline and every other outlet covering it.
