Person
mdnazib963
An independent developer and researcher with an MIT background, focused on large-model inference optimization, who trained a 1.9MB micro language model using cache-fit as the core approach.
Timeline (last 90 days)
Oct 12
The author mdnazib963 trained a 1.9 MB model from scratch that reaches an inference speed of 3405 tok/s on a single core of a Ryzen 5 3400G.
Unverified50%