ludwig

v0.17.5 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 5d LLM Frameworks

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

computer-vision data-centric data-science deep machine-learning deeplearning

+11 more

fine-tuning learning llama llama2 llm llm-training machinelearning mistral natural-language natural-language-processing pytorch

Affected surfaces

breaking_upgrade

Summary

AI summary

GPU Docker images now correctly install the CUDA build of PyTorch, fixing silent CPU fallback during GPU training.

Changes in this release

Type	Severity	Summary	CVE
Feature	Low	Updated Docker images to version 0.17.5, including CUDA 12.6 support for GPU variants. Updated Docker images to version 0.17.5, including CUDA 12.6 support for GPU variants. Source: llm_adapter@2026-05-29 Confidence: high	—
Bugfix	Medium	GPU Docker images now install the correct CUDA build of PyTorch (torch==2.12.0+cu126). GPU Docker images now install the correct CUDA build of PyTorch (torch==2.12.0+cu126). Source: llm_adapter@2026-05-29 Confidence: low	—

Full changelog

What's changed

Bug fix: GPU Docker images now install the CUDA build of PyTorch

The GPU Docker images (ludwig-gpu, ludwig-ray-gpu) were incorrectly shipping the CPU build of PyTorch (torch==2.12.0 without a +cu* suffix) despite being GPU images. This meant GPU training silently fell back to CPU.

Root cause (two issues):

The --force-reinstall step used --extra-index-url instead of --index-url. With --extra-index-url, pip checks PyPI first and finds the CPU wheel (torch==2.12.0) there, so it never looks at the PyTorch CUDA index.
The CUDA index suffix was cu124, but torch==2.12.0 is not published on the cu124 index (which only goes up to 2.6.0+cu124). The fix switches to cu126, where torch==2.12.0+cu126 is available.

Fix:

Changed --extra-index-url → --index-url on the force-reinstall step in both GPU Dockerfiles so pip goes exclusively to the PyTorch wheel server.
Changed cu124 → cu126 throughout both GPU Dockerfiles (including the Ray base image tag).

Verified locally: torch==2.12.0+cu126 with CUDA build version: 12.6 confirmed inside the rebuilt image.

Updated Docker images (0.17.5)

ludwigai/ludwig:0.17.5 — CPU
ludwigai/ludwig-gpu:0.17.5 — CUDA 12.6 (torch 2.12.0+cu126)
ludwigai/ludwig-ray:0.17.5 — CPU + Ray
ludwigai/ludwig-ray-gpu:0.17.5 — CUDA 12.6 + Ray (torch 2.12.0+cu126)

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track ludwig

Get notified when new releases ship.

About ludwig

Low-code framework for building custom LLMs, neural networks, and other AI models

All releases →