x

Teortaxes ️ (DeepSeek 推特 铁粉 2023 – ∞) on X: "@Sentdex depends on price DS 4.1 Flash is larger than GLM 5, but vastly cheaper and faster"

Unless it beats fable/astra, being 2.5T+ would be a mistake imo. We love these models because they are smart, but also bc fast and cheap! Idc anymore about astra/fable...etc bc local ai is so damn ...
circuitdigest.com

Instead of Training a Robot, He Let a General-Purpose Multimodal LLM Control It to Pick Up and Bin a Toy

Harrison Kinsley, the YouTuber behind Sentdex, has shown a Vulcan Robotics Sourccey, a 3D-printed robot he calls “Chonk,” picking up a toy from the floor and dropping it in a bin with no task-specific ...
x

Conor on X: "@Sentdex 4x dgx sparks runs glm 5.3 heavy at 70+ t/s multistream"

https://t.co/7P57BYYR0r We must separate technical competency and GPU ownership from deciding who can do what ...