The updated model runs for free on OpenCode.
new 4 bit quantization of north mini code. you can run this on anything with 20 gigs of memory, or for free on @opencode !
We just dropped three updates that make North Mini Code more accessible than ever ⬇️
No Digg Deeper questions have been answered for this story yet.
North Mini Code is now supported on @ollama. Now, run our model locally, and use it with Codex, OpenClaw, and more. Download it here: https://cohere.link/SVwVbGZ
The 4-bit quant is now available, making our first open-source agentic coding model small enough to be run on a Mac. Get the weights here: https://cohere.link/FjIq0uD
We just dropped three updates that make North Mini Code more accessible than ever ⬇️
@nickfrosst @opencode Interesting feeling to see one of your tweets while listening to you singing x)
Very cool that the model fits on a consumer gpu
You can also use North Mini Code through the @OpenRouter API for free. Play around with weights and make it your own. Weights: https://cohere.link/9tsaUH8
@cohere Great job!
@cohere Make it runnable on 24 gb macbook pro
@cohere Cool.. would like to try it
@cohere @yacineMTB yall kno im canada down 🇨🇦
@cohere @yacineMTB WE NEED YOU FOR CODING, PLEASE
@cohere @OpenRouter 👀
I’ll have to try this out!
@nickfrosst @opencode a cohere model I can run locally! epic :D
@nickfrosst @opencode
You can run it on a lot less than that, and at half decent speeds too 😉 Great release as usual 🫶🏽
@cohere 🤣
@cohere the privacy pitch writes itself: your code never leaves the laptop
Interesting. The local Mac angle is the real hook here.
How should developers think about North Mini Code vs cloud-native coding agents like Cursor Composer?
Is the main advantage privacy/cost/local control, or are you seeing competitive performance on real multi-file repo edits too?
@cohere Can you also release an optiq quant? http://mlx-optiq.com
@cohere Yooo I’m so excited to finally have a Canadian model to play around with ❤️
@cohere keep it up fellas
@cohere @OpenRouter NICE BROO
