Felt the same until I started using Luna. I feel like I get similar performance, but faster, and my usage lasts so much longer.

Yup, GPT-5.6 Luna was a checkpoint for me. Too good for its price.

Someone pays for the AI? That's the new human maintainer.

Who will pay if someone, somewhere is not passionate about it?

You can spin up a model locally and pay yourself. Who will maintain the project if the passionate sole maintainer burns out?

I strictly use only local models, so I agree, but the project was built with Fable so my argument hinges on the assumption that the maintainer is going to continue using Fable and needs to pay for it.

Which model works well?

Qwen 3.6 27b at q6 works decently well on 24GB of VRAM

Hypothetically, maybe. In practice, probably not.

Plenty of important things have been born of passion without necessity.