Grok-4.7 first test results are in and it's not looking good.
Floating island with pagoda in three.js. Bad compared to DeepSeek-v4.1-flash result i got. What makes this even worse is that, this is not one-shot, this was done from Grok build, a complete harness, few turns and
Holy sh*t. This is insane.
Everyone in tech is cheering for America to "win the AI race" against China, but what if America is the bad guy in this story? If a handful of American companies end up controlling the most advanced AI in the world, we’re actually cooked. Think about
Dr. Bratin Saha CEO, NTT DATA AIVista called this exact shift on the Ctrl Alt Podcast!
🚀 CtrlAltpod He foresaw open-weight models like deepseek_ai and Alibaba_Qwen driving the bulk of token volume because "no single model or closed lab will own all the innovation. When
I ran DeepSeek Harness and Hermes side by side.
They look similar until you actually make them work. DeepSeek Harness was better for: → Heavy coding. → Building interfaces. → Faster output. → Easy model swapping. → Cleaner browser-based use. Hermes was better for
You have reached the end of the archive
All of deepseek