Discussion (4 Comments)Read Original on HackerNews

schopra909•about 4 hours ago

Author here, feel free to drop questions below. Will try to answer to best of my ability!

E-Reverance•18 minutes ago

Slightly off-topic but thoughts on this https://arxiv.org/pdf/2410.08159

popalchemist•about 1 hour ago

Does 3.6x faster also = 3.6x cheaper?

schopra909•about 1 hour ago

Yes, 3.6x cheaper when it comes to training!

When it comes to inference it should also be cheaper (since we’ve cut attention sequence length by 4x), but I don’t have a hard number for you how much cheaper for inference

Training Text-to-Image Models 3.6× Faster

⚡ Community Insights

Discussion (4 Comments)Read Original on HackerNews