LTX 2.3 Lora training on Runpod (PyTorch template)

After using the old LTX2 Lora’s for a while with the new model I can safely say they completely ruined the results compared to the one I actually trained on the new model. It’s a little bit of trail and error seeing I was very much inexperienced (only trained on ai toolkit up till now) but can confirm it is way better even with my first checkpoints. Happy