I feel like the models work better in the more minimalist harness. It's totally anecdotal but I feel like the models get lost whenever I run them in OpenCode. Sort of like the "10x"-engineer who overengineers everything.
Not anecdotal at all. Context bloat in Opencode pushes models into the dumb zone sooner.
I' ve heard so. But I'd still consider my opinion anecdotal because I haven't done any research into the subject (or measured it myself) really. All I've done is tried both harnesses.