A Slinky Down an Escalator - GPT 6 Astra max vs Claude Fable 5.1 max
- 发布
- 09/05 15:31
- 收录
- 09/06 14:00
- 来源类型
- 开发者社区
- 档位
- 社区
- 信源状态
- 正常
I finally got access to GPT 6 Astra and Claude Fable 5.1. I wanted to see how far we've come. I was inspired by the "pelican riding on a bicycle" test but wanted to push it a bit more in the direction of inter-object physics, so I came up with this idea for a perpetual slinky going down an escalator. Here I've compared GPT 6 Astra max on the left to Claude Fable 5.1 max on the right.
- GPT 6 Astra max was really fast and didn't burn too much of my weekly usage quota.
- Claude Fable 5.1 max used up the entire 5 hr usage window and errored out once crossing the output token max limit.
- GPT 6 Astra max seems too simple? There's also an issue where the slinky crosses into itself which should not be possible. Is this AGI? Maybe?
- Claude Fable 5.1 max looks more real than I expected. Seems like it passes the eye test, unless I missed something.
Let me know if you spot anything or had better ideas for a test. Overall I'd say Claude wins on quality and GPT wins on speed. Maybe I can tune my prompt a bit better for a more reliable result. This was my original prompt btw:
Make me a single HTML file of a rainbow slinky going down an up-escalator forever. No libraries, just canvas and code. The slinky should be a chain of springs, each coil a different color of the rainbow. It starts folded in half like a horseshoe draped over a step. When dropped it flips end-over-end down the steps and because the escalator keeps moving up it tumbles in place and never reaches the bottom. Include a drop button and a reset button.