For too long, users have lived under the software lottery tyranny of fused attention implementations.
No longer.
Introducing FlexAttention, a new PyTorch API allowing for many attention variants to enjoy fused kernels in a few lines of PyTorch.
https://t.co/IXeUS6AkrY
1/10
🔥 Tarik x Finalmouse Giveaway 🔥
My ULX Pro Series drops this Saturday April 20th at 11am PDT. To celebrate I’m giving 7 away.
To enter:
1. Follow @finalmouse and me
2. RT and like this tweet
Winners announced Friday. Good Luck!
7/ TestGen-LLM - uses LLMs to improve existing human-written tests; reports that after an evaluation on Reels and Stories products for Instagram, 75% of TestGen-LLM's test cases were built correctly, 57% passed reliably, and 25% increased coverage.
https://t.co/N1XGMoii7k
We are excited to share Large World Model (LWM), a general-purpose 1M context multimodal autoregressive model. It is trained on a large dataset of diverse long videos and books using RingAttention, and can perform language, image, and video understanding and generation.
I'm Chiitan, Japan's crazy mascot.
I'm the god of advertising who was born to drive out advertising in the world.
I don't have any products or information to recommend, so please watch the video and laugh.
I hope you have a good day today.