They ran the same prompt on GPT5.6 Sol, DSV4.1-Flash and Opus 5, measuring quality and cost of each.
Everyone argues which AI is smarter; almost no one asks how much each answer costs. Quality and cost together decide which one is worth it.
Dario Amodei's Anthropic just merged two job titles into one prompt.
Claude Opus 5 builds a full 3D website from a sentence. 0 design. 0 code. No Figma, no repo, no "three weeks and a quote." One guy sat down and did it live. The whole recording is 12 minutes. What you watch
Someone just called GPT-6 astra's benchmark run "completed multiple felonies," and the scores behind him make the joke land.
The announcement claims Astra as the most capable and best-aligned model released so far, built across years of pretraining, reinforcement learning, and
GPT5.6 Sol, DSV4.1-Flash, and Opus 5 were given the same prompt
Here is the quality and cost for each
You have reached the end of the archive
All of Claude Opus 5