I tried almost all OS models on opencode, none of them is on levels as opus 4.7.

In latest experiment I used opus for implementation plan then used cursor composer 2.5 for execution.

I must say that combo is really good. Main drawback of claude code is that is super slow. So when paired with composer that is super fast it flies.

No one is claiming that OS is as good. They are saying it isn't that far behind SOTA commercial products. So why pay exorbitantly just to get something only a few percent better than the free option?

But there have been very good open source office apps for decades and few enterprises use them, so perhaps this is just the nature of B2B purchasing committees and 'nobody getting fired for buying IBM.'

Because failures compound. My productivity has substantially improved since I switched from open models to a Codex subscription, because it doesn't need hand holding, and it doesn't pull stupid tricks occasionally.

Do more planning yourself, be smart about the context, break down tasks into smaller components, give it more guidance. You can't just lazily prompt it to complete large features autonomously and expect good results.

But if the closed-source models can do this without the additional effort, that's a significant gap, no?

See that's the thing, they can't. Every model needs hand holding and guidance.

some require less hand-holding than others though

No one is trying to argue that OS models are better than Opus 4.7. It's simply that they're good enough and cheaper.

The point is that the price gap is so much larger than the capability gap, that even with the extra compute needed to make up for the lack of capability, you can still come out ahead in terms of amortized $/work done.

Is it really when they are hundreds of times more expensive?

That is the 3-6 month sota-open gap people talk about, a time-window that continues to move as new models are released on both sides.

Do you know what economic trade offs are?

Both implicit and explicit..?

+1 .. just wanted to reiterate that this is the answer. The open models work great if you just do a little more of the design/architectural work up front and organize your work appropriately.

a good harness is supposed to do what you are describing. sonnet on pi.dev is pretty terrible but fast. Claude Code has ridiculous amounts of prompt engineering at system prompt level and sub session spawing combined with low temperature, to provide the predictable results people like. CC screws up and you never see, because the harness auto corrects, while on OSS you see everything, and does not comes with the level of monitoring by default.