romeinaday's comments

romeinaday · 2026-10-27T22:28:31 1793140111

opendisplay - https://github.com/peetzweg/opendisplay - can make a old iMac's screen useful again. In my case, a 2015 27" iMac can be used as a monitor for my modern Mac laptop, connected via ethernet cable.

Tallain · 2026-10-27T23:00:20 1793142020

Very cool, I like this a lot. Will be giving it a try!

worthless-trash · 2026-10-28T00:39:09 1793147949

I've had good success with https://juicycrumb.com/product/docklite/#imac-model

There is a g4 model too there if thats the version you have.

romeinaday · 2026-08-17T08:19:27 1786954767

Can you run this on a 36GB MacBook Pro (M3 Pro)? What would be a good setup? for coding mainly

petu · 2026-08-17T09:08:03 1786957683

You need ~24-26GB for basic setup (17-19GB model + 128K 8 bit context), so you can, but not much memory would be left for doing anything else on that machine. And even then it would run at like 5-10 t/s (due to relatively low memory bandwidth of M3 Pro) and slow prompt processing (couple hundreds of t/s?)

If they end up releasing updated 35B-A3B variant, then it would be much more interesting in generation speed (~50 t/s)

For inference engine/server you have two (main) choices: llama.cpp for platform-agnostic, MLX for Apple-only. They will spin up OpenAI-compatible local server, and you point your agent harness to it.

For llama.cpp this should be reasonable (maybe shrink context to 128K) starting point: https://x.com/ggerganov/status/2088312671196082312

romeinaday · 2026-08-19T14:37:55 1787150275

So you can use llama.cpp and point a harness to it? What harness do you use?

dstryr · 2026-08-17T16:40:08 1786984808

i have this exact same machine and 3.8-27b runs well, albeit slow at about 6 t/s.

i have it complete tasks in the background as i work on other things.

for comparison, my 3090 runs it at about 60 t/s.

romeinaday · 2026-08-19T14:36:26 1787150186

What are you using to run the model and as coding harness?

dstryr · 2026-08-19T14:51:00 1787151060

llama.cpp + pi.dev

romeinaday · 2026-07-23T21:14:19 1784841259

Where are we with scripting Minecraft worlds? I don't mean in a programmatic sense, but in the sense of saying to an AI "make me a worlds with a clearing in a giant forest where skeletons are attacking creepers" and it would make the world?

PcChip · 2026-07-23T21:16:12 1784841372

a world has a seed as well, so when you walked past the generated portion it would start randomly generating from that seed, which wouldn't match the portion that was sculpted

you would have to make the AI search for seeds that generate what you want at the starting point I think

richstokes · 2026-07-23T21:29:32 1784842172

If the AI could understand the save file format for Minecraft, it could manipulate that

shaewest · 2026-07-23T23:06:58 1784848018

There are methods to do custom generation in of itself, so should be possible for AI to write a mod in the same vein as current world generation mods.

rchard2scout · 2026-07-23T22:22:45 1784845365

There are mods that affect world generation, so I guess it should be possible to write a mod that exposes that functionality to an AI agent.

jamilton · 2026-07-23T23:59:24 1784851164

As of about a month ago there’s https://github.com/soapantelope/mindcraft. You can give it a prompt like “A Minecraft world of beautiful flower-filled meadows abruptly changing into swamps.” and get that. As far as I can tell it’s using Minecraft’s existing world generator, the readme notes “As of now, Mindcraft can only generate worlds that are feasible natural overworld biomes in Minecraft 1.21.11. Keep this in mind when thinking of prompts!”.

And I’ve seen people hook up LLMs more directly with some scripting tools to place blocks/spawn entities, they do okay.

aliasxneo · 2026-07-24T02:03:56 1784858636

I play heavily modded Minecraft and have always dreaded endgame where you're often building out massive infrastructure. Would be amazing to have a little embedded ChatGPT that could generate modded structures with a prompt. The factory must grow :)