Image->HTML test for Muse Wonder. It's much worse than Opus 5.5. IMO it's closer to Sonnet or Haiku for this test.
Muse Wonder (high): https://html.non.io/lcars-muse-wonoder/
Opus 5.5 (high): https://html.non.io/lcars-opus-5.5/
Original design: https://diffui.ai/app/canvas/5093e689-1e74-4f26-b632-2a4500f...
I was expecting much more given the benchmark numbers it was putting out. I suspect its vision model isn't mature yet, and that it's been heavily optimized for text at this point in time.
Additional comparisons:
Sonnet 5.5: https://html.non.io/lcars-sonnet-5.5/
Haiku 5.5: https://html.non.io/lcars-haiku-5.5/
I'd put Muse Wonder somewhere between these two.
Thanks for sharing. Is that Wonder or Wonder Full (barf)? Wonder full is supposed to be the one close to Opus 5.5
> Is that Wonder or Wonder Full (barf)?
I think Meta did a wonderful job of naming this model :D
This is just Wonder - it doesn't look like Wonder Full is up on OpenRouter currently.
If anyone has access to Wonder Full I'd love to send over a prompt for this to see how it does.
Your original design link is requiring a sign in to view it?
Ah shit, that's a bug. Hold please, fixing. (Thank you for commenting btw)
Edit: Fixed - should no longer have the sign in screen: https://diffui.ai/app/canvas/5093e689-1e74-4f26-b632-2a4500f...
I mean, Meta themselves put Wonder at much lower on the benchmarks they've released than Wonder Full. So this is not very surprising from base Wonder.