if you want to try a diffusion LLM, inception labs just released one. it goes 1000 tokens/sec on regular H100s, absolutely nuts how fast it is
speed alone makes these things interesting
www.inceptionlabs.ai/news
if you want to try a diffusion LLM, inception labs just released one. it goes...
View original thread
@timkellogg.me
LLMs That Don't Gaslight You
A new language model uses diffusion instead of next-token prediction. That means the text it can back out of a hallucination before it commits. This is a big win for areas like law & contracts, where global consistency is valued
timkellogg.me/blog/2025/02...
A new language model uses diffusion instead of next-token prediction. That means the text it can back out of a hallucination before it commits. This is a big win for areas like law & contracts, where global consistency is valued
timkellogg.me/blog/2025/02...
61
7
10 hours later
seems like a promising application of dLLMs is real-time systems, where you have a strict upper bound of how long things can take, e.g. robotics, cars, etc.
4