Self-hosting Kimi K3: 20% more hardware cost, 20% better task resolution (aistack.imec-int.com) | HN Screenshots

YHacker Newsscreenshots

Screenshot of aistack.imec-int.com: Self-hosting Kimi K3: 20% more hardware cost, 20% better task resolution

First screen, captured 2 months ago.

Looks similar

Screenshot of injuly.in: Inference cost at scale with napkin math

Inference cost at scale with napkin math (injuly.in)

74 points by gmays 4 months ago | 15 comments

Screenshot of devblog.qnx.com: QNX Self-Hosted Developer Desktop

QNX Self-Hosted Developer Desktop (devblog.qnx.com)

257 points by transpute 10 months ago | 147 comments

Screenshot of baseten.co: Running GPT-OSS-120B at 500 tokens per second on Nvidia GPUs

Running GPT-OSS-120B at 500 tokens per second on Nvidia GPUs (baseten.co)

215 points by philipkiely 1 year ago | 142 comments

Screenshot of indianspeedster.github.io: Occupancy Math on the AMD MI355X: A From-First-Principles Guide

Occupancy Math on the AMD MI355X: A From-First-Principles Guide (indianspeedster.github.io)

43 points by skidrow 4 months ago | 4 comments

Screenshot of thegeeko.me: AMD GPU Debugger

AMD GPU Debugger (thegeeko.me)

268 points by ibobev 10 months ago | 49 comments

Screenshot of piszczek.pl: Qwen3.8-27B at 256K on a 24GB RTX PRO 4000 SFF (432 GB/s): 50 tok/s with MTP

Qwen3.8-27B at 256K on a 24GB RTX PRO 4000 SFF (432 GB/s): 50 tok/s with MTP (piszczek.pl)

31 points by pich 2 months ago | 24 comments