KC

Design & Code.

[KC @kcwolfy_]

Codex mode UI now pretty much exactly the same as Claude. Two modes: Home/Work and Coding. Place your bet on when two goes to one and the first to get there?

UX (inc UI and other modalities e.g. voice) should wrap itself around the user's intent. You only need a single starting point, a place to begin that adaptation.

[KC @kcwolfy_]

@CSProfKGD Agincourt.

[KC @kcwolfy_]

Uploading these two quantizations of
Qwen3.8-Flash-Next for oMLX to @huggingface .

https://t.co/5aACHzw8Sv
Tested with: Apple M3 Ultra, 256 GB · oMLX 0.6.4 · thinking off · temperature 0 · Lightning MTP depth 3

[KC @kcwolfy_]

@dhh ... and don't forget to talk to https://t.co/ZkxU1ogrkw (Crossover). Thanks.

[KC @kcwolfy_]

@dhh THIS!!

[KC @kcwolfy_]

@Polymarket ... and will be the moment where the Republican Party finally gets the balls to turn their back on Trump.

[KC @kcwolfy_]

Things nobody in AI has ever said:
“That’s probably enough context.”

[KC @kcwolfy_]

@MAGACult2 AI slop. She does not breath.

[KC @kcwolfy_]

@bilawalsidhu Meanwhile... back at the ranch.

[KC @kcwolfy_]

@KapoorShan @elonmusk @rauchg Right!?

[KC @kcwolfy_]

@WordsnWisdom26 As every Republican knows.

[KC @kcwolfy_]

@weswinder Because X does it for them?

[KC @kcwolfy_]

@Enezator It's a dog, you LLM.

[KC @kcwolfy_]

@Mdw37683404 @BNONews Thank you for your comments, Moscow Rage Beta.

[KC @kcwolfy_]

@thinkingwest That's incorrect. Mathematically, I'm pretty sure. Mathematicians step in. It will just run exceptionally slowly. By which I mean hundreds of millions of years, probably, to even get the first frame on the screen. Anyone want to do the math on this?

[KC @kcwolfy_]

@caty_komal Me. Vibecoding. Time: 3 months. Cost: $250,000K. Any takers?

[KC @kcwolfy_]

@dhh Great to hear! And congratulations. I was concerned for a moment it was gonna be some weird base camp offshoot, and it just didn't feel like it fit very well.

[KC @kcwolfy_]

That's a large oversimplification. When you look at the growth of use for them.

What we can say is that very good models are getting very close, and they're much cheaper.

I'm really interested to see with the next model releases from Anthropic: do they try and maintain the same level of price premium?

[KC @kcwolfy_]

@WescheNex1q My search continues. I don't want to run a full deep swe bench either, but maybe I'll run a subset. If you have any thoughts, let me know.