Hacker News Reader

Comment by rspeele | Hacker News Reader

On the Claude side of things I was previously following "strong model directs weak" with Fable directing Opus/Sonnet (its choice per-task). Since Opus 5.5 came out I've just been having Opus direct Opus.

The sub-agent separation is still valuable to keep context clean for the orchestrator, but I just have no reason to use Sonnet as the grunt-work implementer because I'm finding it hard to run out of tokens with Opus 5.5 on a $200 subscription plan. It's really really good at subjective quality of work per token used.

View on HN

I would probably go that route if I could use other harnesses with claude models, but I don't want to be locked in to claude code, and their API pricing (which you need to use it with other harnesses) is so much higher than subscription.

plenty of harnesses use the claude subscription

Ever since Claude banned this, I haven't found one. Which harnesses have you found that allows using Claude through the subscription plan?

oh my pi has it.

I just checked, it has claude available through the API. I asked Claude itself and it said the subscription service is only available through Claude code.

Did you use some kind of plugin? I'd like to use opencode with my Claude subscription.

pi.dev has it. https://pi.dev/packages/pi-claude-code-provider is the plugin you need to install and everything goes through claude subscription

Thanks for the link, though that particular project is abandonware.

The active project seems to use Claude code's agent API https://github.com/anthropics/claude-agent-sdk-typescript so yes you can issue commands to Claude code from opencode, but the harness in that case is still ultimately Claude code, with all the big fat system prompts that come with it.

My work pays for Claude and cursor.

I have actually just dropped to using sonnet for everything, sure it does need some directing but I have yet to see a need to jump to opus.

To me it feels like sonnet/terra and composer 2.5 and grok 4.7 are actually good enough for most tasks and these companies are pushing the high models simply to make money.

Every test I've done with Sonnet or Tera has been more expensive than just using their big brothers, even before 6.0 and 6.1 Sol came out at Tera prices.

They're cheaper per token, but they burn more tokens and take more wall time than just going up a tier at a lower reasoning level. And if the task is simple enough that they do come out cheaper than Sol, I can switch to Luna and come in even cheaper.

Actually not tested that I will give it a go. I usually keep reasoning on low.

Reply on news.ycombinator.com