GLM-5.3-Flash vs. DeepSeek-v4-Flash
Same prompt, same harness (pi), same effort. GLM used 114k tokens over 93 turns taking 73min. Total cost 16 cents. DeepSeek used 70k tokens over 9 turns taking 10min. Total cost 2 cents. GLM is a richer scene but at 8x the cost and 7x the
3.8 Flash Next runs perfectly on my Android phone (12GB RAM)
As you know, my Bigmoeonedge project enables running massive models on edge devices - such as a mid-range Android phone with 12GB of RAM. Following DeepSeek and various other models, Qwen 3.8 Flash Next
12 new AI models dropped in the first three weeks of august.
From 7 different companies. most of them you installed, tested for 20 minutes, and never opened again. this is what AI feels like in 2026. not one breakthrough — twelve. not one choice — twelve. GPT-5.6 Luna. Gemini
This video is a reaction to Sam Altman's interview, and the thesis is blunt
"there is no secret plan, Altman made a massive bet and lost, there is no way to save OpenAI." That's a strong claim, and clearly an opinion — but the numbers backing it up are real and verifiable.
You have reached the end of the archive
All of deepseek