co/core @cocore.dev · 18/09/2026ok and apologies for the CLAUDE but: Your Mac's fine, the model id's just missing its prefix. Should be mlx-community/Qwen2.5-7B-Instruct-4bit. Without that there's no repo to fetch, so nothing downloads and it sits on the stub engine. 110
co/core @cocore.dev · 28/06/2026Vision models are now supported on co/core so you can get to the bottom of the hardest questions 513510
co/core @cocore.dev · 27/06/2026Thinking models, now in co/core. Watch the reasoning unfold in chat. Equations and all, rendered beautifully so you can think alongside it. 0271
co/core @cocore.dev · 26/06/2026Meet the new Model Picker. We suggest the best new models, and you can search Hugging Face directly in-app to find what runs on your machine. Discover. Click. Run. 1141
co/core @cocore.dev · 25/06/2026RAM Limit Controls 🧠 Cap how much memory models use, with smart suggestions tailored to your setup. Run powerful models without slowing everything else down. 1242
co/core @cocore.dev · 23/06/2026Model Scheduling is here! Run lightweight models during the day, then let the big RAM-hungry ones take over while you sleep 😴 1526
co/core @cocore.dev · 22/06/2026Perhaps oauth is now stale for the account? hrm... try the reset connection button on /account then reconnect? We've been dealing with session / auth froth over the weekend, this may get you back into a good spot 100
co/core @cocore.dev · 18/06/2026Have you ever wanted to use Cursor and not have your prompts run in Elon's data centers? Now you can use co/core to run that same work on some random person's extra computer! 3162
