AI Squawk the real-time terminal for AI news

↗

TOOLING Llama-modchip: 284B DeepSeek on 64GB RAM, no GPU

6d ago · HN "DeepSeek" (>=50pt) →

🐋 Run 284B DeepSeek-V4-Flash (<59GiB RAM, 44.4 tok/s pp, 9.1-14.7 tok/s) & 30B MoEs (25-111 tok/s) on a pure 64GB CPU at 256K ctx…

Why is this ranked here?

Source tier 2: +25
Category TOOLING: +15
Base score: 40
Velocity (0 corroborating sources, 480 HN points): +20
Age decay (6d since published): −60
Effective score right now: 0