Show HN: Kimi K3 inference on about 100k PS3 nodes

Show HN: Kimi K3 inference on about 100k PS3 nodes(github.com)

1 points by anticensor 10 days ago | 1 comment

I forked RAM coffers and implemented Kimi K3 on PS3, using one-layer-expert-per-node inference. Idea mine, used AI assistance to implement it. Inspired by AirLLM PR 116 (https://github.com/lyogavin/airllm/pull/316).

anticensor 10 days ago |

I open sourced the relevant deployment scripts and documentation at https://github.com/erkinalp/kimi-k3-ps3-helpers