Tag

Inference

2 articles

Google's Gemma 4 Just Got 3x Faster - MTP Drafters Land for the Open Models

NEWS

Gemma 4, Now 3x, Faster!

0:52
play_arrow
FPGA MicroGPT Hits 50,000 Tokens/Sec - A Transformer With No GPU, No PyTorch, No CPU

NEWS

50,000 Tokens, Per Second, MicroGPT On Silicon

0:55
play_arrow

Related

open source inference optimization speculative decoding edge AI FPGA hardware acceleration transformer
Back To News