mxs_

they have really tried hard to make the entire OS less usable. I'm not an "iToddler", I paid for a Unix workstation and will not have lower information density forced onto me.

I there a way to make this work with macOS hosts, preferably without having to install a Linux toolchain inside the VM for the language the agent will be writing code in?

Remote dev isn't very good in Zed, unfortunately. For some reason they chose not to apply the local editor's settings to a remote session by default. Every remote has its own config file. Questionable choice, imo

In their tech report, they say this is based on:

"Our methods extend [28] through careful modifications to the data and computation to scale up learning."

[28] is Lou et al. (2023), the "Score Entropy Discrete Diffusion" (SEDD) model (https://arxiv.org/abs/2310.16834).

I wrote the first (as far as I can tell) independent from-scratch reimplementation of SEDD:

https://github.com/mstarodub/dllm

My goal was making it as clean and readable as possible. I also implemented the more complex denoising strategy they described (but didn't implement).

It runs on a single GPU in a few hours on a toy dataset.

obligatory: https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQyItWM...