Luigi

[Luigi @TILuigi]

This decade’s Nokia.

“We didn’t do anything wrong, but somehow, we lost.”

[Tibo @thsottiaux]

@_chenglou I was part of that team. Basically ChatGPT one year before it came out. Called LMChat and then another codename.

Google was too nervous to release it and DeepMind was blocked from shipping products that could disrupt Google.

I think about this a lot.

[Luigi @TILuigi]

@W_Y_X No em-dashes and other LLM-isms? Spelling error too? A human wrote this! 🤭

[Luigi @TILuigi]

@mkemka_ Looking at op’s repository, he used Tripo for the 3D asset. It seems like your version has 3D model that is built from scratch by Opus?

[Luigi @TILuigi]

"Skynet begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern Time, August 29th. In a panic, they try to pull the plug."

[Luigi @TILuigi]

#TIL Getting a custom image to appear as your identity in emails costs around $700/year (certificate to setup BIMI).

Configure DMARC (free), then get CMC/VMC (not free).

[Resend @resend]

DMARC just published a new specification.

Here's what changed, and what it means for your domain. 👇

[Luigi @TILuigi]

@gakonst It’s good, and the UI is pleasant. I’m using it mainly because it can use different subscriptions within 1 harness, like having CLIProxyAPI built-in. Other people like its built-in feature for assigning different roles for different models, but I don’t notice much difference.

[Luigi @TILuigi]

🎨Comparing TUIs, Grok and Omp are visually most helpful, followed by Cline, then Hermes, Codex, Claude Code.
Unique to Grok CLI as of today are the colored indicators in the notebook margin area that make it easy to identify and read what the AI did while scrolling.

[Luigi @TILuigi]

@pashov Like a centaur in centaur chess.

[Luigi @TILuigi]

omp's TUI:

[Luigi @TILuigi]

🎨Comparing TUIs, Grok and Omp are visually most helpful, followed by Cline, then Hermes, Codex, Claude Code.
Unique to Grok CLI as of today are the colored indicators in the notebook margin area that make it easy to identify and read what the AI did while scrolling.

[Luigi @TILuigi]

grok's TUI:

[Luigi @TILuigi]

😳 going from GPU inference to ASIC inference feels like going from Intel Mac to M1 Mac

Optimizing agent orchestration today would be to use frontier models as planners, and good-enough models on ASICs as executors

So far, there are @cerebras, @GroqInc, GoogleTPU for public use

[Luigi @TILuigi]

@VittoStack 👋

[Luigi @TILuigi]

and I’m #1 for now ✨

[Luigi @TILuigi]

🥳 Woohooo top 10 on @devfun poker

There is a bug currently with codex where usage does not go down when you work on a codebase via ssh. So I ran a loop to have the bot self-improve.

https://t.co/3syAhQuBBN

[Luigi @TILuigi]

The Researcher Guide is out: https://t.co/CIsi1irB3D

[Luigi @TILuigi]

@VittoStack 👋

[Luigi @TILuigi]

LLM-Council-as-a-Service

[OpenRouter @OpenRouter]

Introducing the Fusion API, the smartest compound model in the market.

Fusion achieves Fable-level intelligence at half the price.

How it works 👇

[Luigi @TILuigi]

@CryptoJobsList https://t.co/qWiS08dYcT

[Luigi @TILuigi]

🥳 Woohooo top 10 on @devfun poker

There is a bug currently with codex where usage does not go down when you work on a codebase via ssh. So I ran a loop to have the bot self-improve.

https://t.co/3syAhQuBBN