Cool I hope to try it someday when I can run a local model good enough for coding. Until then I will probably stay with opencode for the time being.

Local Qwen 3.5 on CPU was good enough for small tasks.

You can use it with oai subscriptions

don't worry you are popular with me

Imagine Claude 5.6 starts saying things like “dare I say… load bearing” it wouldn’t be cute anymore

It’s just poor ux on mobile

Reading the article, this does seem ill-advised.

Open AI admits to such here: Open AI aims to have a stable API and admits to meddling with effort levels and such for subscriptions -https://news.ycombinator.com/item?id=49804316#49809266

Open AI aims to have a stable API and admits to meddling with effort levels and such for subscriptions here: https://news.ycombinator.com/item?id=49804316#49809266

I think most people are on 'auto' mode nowadays.

I am skeptical that individuals on the $200 and $500 plans make up that meaningful of a portion of revenue.

They use a meaningful portion of compute.

The way I use openrouter is I find a model/provider combination I like then pin all requests for that model to that single provider.

If you disable all other providers but DeepSeek in your OpenRouter guardrails, is that effectively the same thing?

Sure but you still pay only OpenRouter :)

Interestingly, OpenRouter will serve inference at cost for a lot of models.

No? They charge 5.5% (I believe) on every top-up. Which is the equivalent of charging you 5.5% on every request.

Ah, so that's where the profit comes from. Very payment-processor like move, makes sense that Stripe picked them up.

How do you do this?

provider: { order: ['deepinfra/turbo'], allowFallbacks: false, },

https://openrouter.ai/docs/guides/routing/provider-selection

What harnesses are capable of doing this provider pinning when doing requests on openrouter? Can DeepSeek harness (dsh) do it? Or opencode etc

I stopped using opencode because it has some issues with caching, so I suppose it doesn't do this

Deepseek plugins* all the way down. It should be easy to add support for this.

Using this interesting framework called Cordis I've recent discovered.

Open code and pi can do it. Unironically, I let Claude configure these things for me.

What pests said. And you can make a preset and pass "model": "@preset/deep-seek"