I've also build a solution that uses llm council, loops, context engineering, etc, fully open source, if you're interested - https://github.com/looptroop-ai/LoopTroop
Such a low quality post. I went to buy a chuwing gum and because they charged 10% more than last week I decided not to. And decided to write a blog post about it. And a Ted talk. Very interesting
Very good models from China, minimax, ernie, glm, deepseek, mimi, Qwen. Hope I didn't forget anyone else. Future looking good for AI model competition. (or bad for humans, depending how you see it)