Still thinking about the @gensynai community meetup we had in Bengaluru on 4th December.
What started as a small idea turned into one of the most energising community meetups I’ve been part of. Bengaluru showed up with so much heart 🫶
When you run the same AI model with the same inputs twice, you'd expect the same output.
But modern GPU execution is optimised for speed, not fixed ordering, and existing determinism tools do not solve this across hardware.
Today we're changing that.
https://t.co/wIVL5mCvBO
@GadgetLeo there was a refund policy available btw.
you could have chosen to refund your allocation if you weren’t happy with it. pretty risk-free imo :)