Reactions from ranked influencers
4 posts
Lol, the text diffusion research team at DeepMind has existed before Inception was even founded. Also totally bizarre to compare a 1T parameter closed source model to a 30B open weights model 🙃
Welcome to the diffusion era. We bet on parallel generation years ago, when it was a contrarian idea. It's great to see the industry arrive. Mercury 2 continues to lead the Pareto frontier for quality, speed, and cost among publicly available diffusion LLMs.
Lol, the text diffusion research team at DeepMind has existed before Inception was even founded. Also bizarre to compare a 1T parameter closed source model to a 26B open weights model 🙃
Welcome to the diffusion era. We bet on parallel generation years ago, when it was a contrarian idea. It's great to see the industry arrive. Mercury 2 continues to lead the Pareto frontier for quality, speed, and cost among publicly available diffusion LLMs.
@bodonoghue85 Curious what do you think is missing from mainstream adoptions (what's hidden interms of scientific questions to answer before it becomes practical)
@_arohan_ These issues can be solved though with more research, part of the reason we open-weights released DiffusionGemma is we wanted to do something similar to how early Llama models were great base models for research and helped to push the field a lot.