News for hackers. Stuff that matters.
Re:Openrouter Fusion API by qsort on Monday June 15, 2026 @06:34AM (#48539301)
Yeah, GPT 5.5 + Fable beating either individually is belivable, but 2x Opus > Fable is what makes me a bit dubious about the whole thing. They might be measuring skills that are too specific or benefit a lot from more tokens being thrown at them. Also Claude Code (the harness) is not the best at the moment, that might be part of it as well?
Re:Openrouter Fusion API by andai on Monday June 15, 2026 @10:18AM (#48541676)
What throws me off is DeepSeek beating both Opus 4.8 and GPT 5.5.

That definitely doesn't sound right.