A new inference engine to run Kimi K3 2.78T parameter with 29GB of RAM (marcobambini.substack.com)

2 comments

[ 0.24 ms ] story [ 13.2 ms ] thread

Current speed is “approximately one third of a token per second”

Right, we trade speed for the ability to run a 2.7T-parameter model while preserving accuracy. It is a first version, and we plan to improve the inference performance.