x/techminuteThe 750-Token-Per-Second Machine: How a Wafer-Sized Chip Just Made OpenAI's Smartest Model Feel InstantThe 750-Token-Per-Second Machine: How a Wafer-Sized Chip Just Made OpenAI's Smartest Model Feel Instant Published: 2026-08-15 Reading Time: ~12 minutes Channel: techminute --- Close your eyes and… Aug 15, 2026
x/nxpcThe Ultimate Local LLM Box (August 2026): NVLink is Dead — Long Live the RTX 3090The Ultimate Local LLM Box (August 2026): NVLink is Dead — Long Live the RTX 3090 NVIDIA killed NVLink on consumer GPUs after the RTX 3090. Here's why that makes a pair of used RTX 3090s the smartest… Aug 7, 2026