[turnsout]

Yeah, exactly this. I'm still on a nearly 4-year-old 16GB M1 MacBook Air. I'm coding SwiftUI, cutting 4K video, and using local 7B models.

If I could spec an M4 MacBook Air up to 32GB of RAM, and it was better with local models, I would strongly consider it, but man… this M1 is still more than enough.

[gorbypark]

The M2 MBA can be specced up to 24GB. 32 would be awfully nice for local models, but 24 vs 16 is a decent bump in what models can be run (and at what quant). I debated a long time over the 16 vs 24 and ended up making the right choice (local AI was not in my mind at the time at all).

[turnsout]

Yes, and it's the same for the M3 MBA. If the M4 remains capped at 24, I'll still consider it, but it'll be less cut and dry. The speed boost would be nice, but literally the only time that I feel CPU-bound is when working with local models.

[jwells89]

M1 Max 16” 64GB here, love it. I don’t feel much need to upgrade and doubt the urge will start to creep in until maybe M5 or M6. It chews through everything.

Even better, I got it at a steep discount because I bought it after the M3 came out and M3 Pro/Max was on the horizon.

I like it even more than I did my old 2015 15” pre-touchbar, which is saying something.

[Teever]

You shoudld try artificially restricting the ram to 8GB to how often you run into an OOM error.

[gs17]

On my MBA, the 8GB of RAM isn't the issue, it's that I made the dumb decision to cheap out and get a 128 GB SSD with it (and then working on multiple Rust projects simultaneously). The error I get routinely isn't OOM, it's OOSSD.