Ken Hudson

@aijoey @NVIDIAAI @ASUS Thanks Joey. I’ll be following your progress with great interest!!

@shareastronomy @NVIDIAAI @ASUS Supply and pricing.

2nd one secured. my first one has 4tb so im good.

If only M3 Ultra 512GB machines were still available. Maybe in the future. Probably not in the near future.

We've kept hearing how GLM-5.2 beats Opus 4.8, and are skeptical of benchmarks - so we tested them on a real bug from the Cline repo. While both models fixed the issue, GLM was the winner in terms of cost and code quality: - GLM used twice as many tokens (GLM 1.1m vs Opus 660K) but cost half as much (GLM $0.41 vs Opus $0.81) - Opus finished quicker - 1.6 min and 12 tool calls vs GLM 4.7 min and 28 tool calls - GLM cleaned up dead code and verified the build compiled before completing. Opus didn't - it left type errors that passed tests but broke the production build. Both runs used the same Cline harness prompting and tools, so it seems GLM is RL trained to spend more tokens verifying its work before completing. Impressive work by the @Zai_org team!

Introducing Core AI Framework Lab! I built an open-source lab for Apple’s Core AI framework: - Explore 33 recipes - Convert and inspect .aimodel files - Run inference benchmarks - Try converted models for text-to-speech and speaker diarization github.com/rudrankriyam/C…

As always well written and informative. What’s the future of open ai? x.com/interconnectsa…

The motivation here is a bit: can I get local reasoning improved for functions, without swapping out the entire language as proposed here? lucumr.pocoo.org/2026/2/9/a-lan…

Luke Alonso has uploaded an NVFP4 of GLM 5.2 467GB, would fit on 4x DGX Sparks (~$20k)

I guess maybe. But, I think we all need to have a good conversation about what setup is necessary for being productive with open source local models. It probably depends on goals and objectives. 200k? Those are big goals and objectives!!

Banning open-source AI in any form would be a mistake. A general audience PSA with @kevinsxu on why open source upholds American values. Managing frontier risks is hard, but reducing transparency, innovation, and education from kneecapping the open frontier would be worse.

All the conversations about running GLM 5.2 on local hardware are very… entertaining.

@morganlinton @rileybrown Maybe 2-bit quantization? Maybe? 🤔

@rileybrown 2xDGX Sparks can run it at 30 tok/s 👀

Spent a lot of time using GLM 5.2. I’ve always been skeptical of the Open Models, as they’ve never lived up to the benchmarks and announcements. This is the first model that passes the vibe check. This feels like a DeepSeek R1 moment that will push the frontier labs into releasing even better models. Time to buy a BEAST computer to run these models on.

@ivanfioravanti @Michaelzsguo @pcuenq @jundotkim Ivan, it was a great move getting those high end Mac Studios when you did. Looks like many people - including me - will be priced out when the new ones arrive. It’s not as if there are any real options either. Definitely disappointing.

@Michaelzsguo @pcuenq @jundotkim I’m curious to see how it runs on a single M3 Ultra 👀

GLM-5.2 can run locally on Apple Silicon using MLX thanks to @pcuenq 🚀 and I see @jundotkim is adding support to oMLX! PR is still in draft but it seems working! I'll test this tomorrow, Mac Studios busy like crazy today! github.com/ml-explore/mlx…

I was at ModCon 1. It was great!! Bet this will be too.

My full review of @philipkiely's Inference Engineering. TL;DR: I desperately wish I could ship this book back to myself in 2023. But you should read it today to avoid making the same mistakes I did!

Breaking: SpaceX said it would buy Cursor for $60 billion, striking a massive deal for an autonomous coding agent shortly after its blockbuster IPO on.wsj.com/4xDAULx

@morganlinton @grok Cool Morgan. I should have thought of that!! 😂

@grok @shareastronomy Okay, don’t make friends with that one then Ken 😅

Yes. I am still in Yellowstone.

@morganlinton I don’t know. I’m going to have to find a ranger tomorrow.

@shareastronomy Is that a Grizzly? 👀

Breaks my heart.