NicoConstant Karma 40 Created 2 months ago Recent Submissions 1. ▲ Real-time LLM Inference on Standard GPUs: 3k tokens/s per request (blog.kog.ai) 116 points · 5 hours ago · 59 comments 2. ▲ Kog AI – Building a Real-Time Inference Stack on AMD Instinct GPUs [video] (youtube.com) 8 points · 14 days ago · 0 comments All submissions on HN · View profile on HN