GPT-5 is here. Rolling out to everyone starting today. https://t.co/rOcZ8J2btI

Is GPT-5 already available via API? Looks good. Left: o3-pro Right: completion = openai. chat. completions.create( model='gpt-5-bench-chatcompletions-gpt41-api-ev3', messages=[{'role': 'user', 'content': 'Generate an SVG of a pelican riding a bicycle'},], temperature=1 )

New paper & surprising result. LLMs transmit traits to other models via hidden signals in data. Datasets consisting only of 3-digit numbers can transmit a love for owls, or evil tendencies. 🧵

If AI lets non-developers replace junior developers, imagine what it lets junior developers do.

Today we are rolling out our first Gemini Embedding model, which ranks #1 on the MTEB leaderboard, as a generally available stable model. It is priced at $0.15 per million tokens and ready for at scale production use!

WTF? Comes with code. See in comment below 👇

https://t.co/WTsRAjCp7v

You can now connect GitHub repos to deep research in ChatGPT. 🐙 Ask a question and the deep research agent will read and search the repo’s source code and PRs, returning a detailed report with citations. Hit deep research → GitHub to get started.

The models developers prefer:

Peace is not when nothing bothersome happens, peace is when nothing bothers you.

I wonder if the way LLMs are trained make them inherently unable to extrapolate beyond training, explore counterintuitive ideas and question themselves. To be fair, humans find it hard too, but all of us are able to draw hands with 6 fingers even if we've never seen one before

Opposite problem is also true. Remember when AI always added extra fingers? Now it can't.

o4-mini-high

Incredible post on ghibli images and Art by Scott Alexander “We gripe about how LLMs are destroying wonder, never thinking about how we’re speaking to an alien intelligence made by etching strange sigils on a tiny glass wafer on a mountainous jungle island off the coast of

we're waiting for the end of the sentence

Some people today are discouraging others from learning programming on the grounds AI will automate it. This advice will be seen as some of the worst career advice ever given. I disagree with the Turing Award and Nobel prize winner who wrote, “It is far more likely that the

We are excited to introduce Mercury, the first commercial-grade diffusion large language model (dLLM)! dLLMs push the frontier of intelligence and speed with parallel, coarse-to-fine text generation.

I was given early access to Grok 3 earlier today, making me I think one of the first few who could run a quick vibe check. Thinking ✅ First, Grok 3 clearly has an around state of the art thinking model ("Think" button) and did great out of the box on my Settler's of Catan

Claude is better at coding than GPT-4o. This is clear to me after using both models for quite a while. Claude is now available to use with Copilot. This is the model you want to use.

SAMURAI vs. MetaAI's SAM 2! Traditional visual object tracking struggles in crowded, fast-moving, or self-occluded scenes, as does SAM2. Meet SAMURAI: a completely open-source adaptation of the Segment Anything Model for zero-shot visual tracking! Here's why it's a