dmlc / dmlc/xgboost

train on CUDA and load on CPU but got an EXC_BAD_ACCESS crash

Open
#11,944 14 comments 0 reactions 0 assignees View on GitHub
Dominant language
C++
Stars
28.8k
Forks
8.9k
Avg merge
1d 12h
Merged PRs (30d)
54

Description

trained my XGBoost model on CUDA, saved in json format and then tried to load model on macOS, got an EXC_BAD_ACCESS crash.
The python versions on both platforms are 3.10.x, and XGBoost versions are both 3.0.3.

Full crash log attached.

[Python-2026-01-21-163640.log](https://github.com/user-attachments/files/24761527/Python-2026-01-21-163640.log)

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.