Hacker News new | ask | show | jobs
user: GreenGames
created: 2020-09-14
karma: 233

submissions:

A 35B MoE on a 16 GB GPU, without the offload tax
18 points | 0 comments
PFlash: 10x prefill speedup over llama.cpp at 128K on a RTX 3090
3 points | 1 comments
0 points | 0 comments
0 points | 0 comments
We got 207 tok/s with Qwen3.5-27B on an RTX 3090
165 points | 52 comments
Show HN: OS Megakernel that match M5 Max Tok/w at 2x the Throughput on RTX 3090
6 points | 1 comments
0 points | 0 comments
Cua (YC X25) is hiring an engineer
1 points | 0 comments
0 points | 0 comments
App-Use, Control Individual Applications with CUA Agents
2 points | 1 comments
Show HN: Lumier – Run macOS VMs in a Docker
159 points | 52 comments
0 points | 0 comments
Microsoft is reportedly about to lay off 3% of its workforce
13 points | 2 comments
Polaris is giving free GPUs/CPUs for everyone
3 points | 0 comments
Improvements in reasoning AI models may slow down soon, analysis finds
3 points | 0 comments
0 points | 0 comments
Microsoft and OpenAI are renegotiating their partnership
3 points | 0 comments
Apple developing new chips for smart glasses, Macs, and more
6 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
Show HN: Tlume – a CLI tool that converts Tart VM images for Lume
5 points | 2 comments