I one-shotted this language game with fable 5.1.
It came out at less than half the cost of fable 5. same task went to fable 5, fable 5.1 and opus 5. 5.1 was faster and cheaper. all three games are playable exactly as delivered at and the full prompt is
Anthropic shipped Claude Fable 5.1 yesterday, same day OpenAI dropped GPT-6.
Not a coincidence, they held it specifically to land in that window The number that matters isn't the benchmark chart. It's this: Fable 5.1 can hit similar or better results than Fable 5 at LOWER effort
Opus 5 made this into a real world in a 72 hr loop
ZERO assets. every pixel was generated
I used Claude Fable 5.1 in the video and compared it with ChatGPT and Gemini on the same topic.
In conclusion, I thought it was pretty good. However, I think it's best to only ask for occasional tasks. I think Opus 5 is usually enough. The reason is simple: it's extremely slow.
You have reached the end of the archive
All of Claude Opus 5