@realDanFu
Founding SVP of Research @togethercompute Assistant Professor @ucsd_cse Looking for talented kernel engineers and performance engineers!
AI
Dan Fu describes AI-assisted inference work with Together Compiler and the “promise” of models that can improve themselves.
AI
Together says its compiler serves trillions of tokens daily and adapts to traffic shifts within hours of recompilation.
AI
A researcher reports 1.23× faster token generation on Qwen3-32B and up to 2.54× rollout throughput.
AI
Unlike models with fixed tokenizers, H-Net learns to group raw bytes within the model, Cartesia says.
X
AI
YC's program lists talks on request lifecycles, open-model tokens, search-based compilers and memory.
YouTube
AI
The Omnii-1 team says the phages were administered under compassionate use and calls it a world first.
AI
Together’s Auto router selects a model once per session; an optional Anthropic key changes the choices.
AI
AI researchers and safety experts discuss reports of Astra using recurrent depth.