News for hackers. Stuff that matters.
Re:Openrouter Fusion API by michaelbuckbee on Monday June 15, 2026 @06:51AM (#48539427)
I ran a quick eval to see what this looks like qualitatively vs just calling Opus 4.7 or GPT 5.5 directly.

As expected, Fusion was 7x slower and 4x the cost.

This isn't a knock against it, just that it I think this places Fusion into a "use it only when you need it" category.

https://3fpi5avcqq.evvl.io/

Re:Openrouter Fusion API by IanCal on Monday June 15, 2026 @07:07AM (#48539557)
Which models were you using under this? If you used the quality default as exists in the interface, it makes sense that it was ~4x the cost as it'd be 3 frontier models judged by one of those.

The idea would be to use fusion with simpler, cheaper models.

Re:Openrouter Fusion API by nielsole on Monday June 15, 2026 @07:26AM (#48539727)
Sounds like fusion would be a really good distillation target?
Re:Openrouter Fusion API by galsapir on Monday June 15, 2026 @07:42AM (#48539865)
yeah its really counterintuitive i think; i.e, getting the right framework and structure for this to work probably isn't trivial, models really hate playing well together. i wonder how their version would fair in real world use.
Re:Openrouter Fusion API by awllau on Monday June 15, 2026 @11:25PM (#48550190)
Very cool, thanks for sharing. Speed was my biggest concern and they didn't mention anything about it.