Still golden

https://www.youtube.com/watch?v=rsFHqpN2bCM

1/12 day of OpenAI

Benchmark don't look too promising to be honest and Sonnet still seems like a king. But who cares about benchmarks, lol

While everyone is talking about fine tuned o1 and continues to giggle about monthly price for o1 pro, Google finally starts releasing some actual good shit

On my own humble set of tests it also showed itself… great

A bit not what I would expect in terms of writing code + it feels like it may hallucinate a little bit too much when there’s a big context but anyway it’s definitely worth a shot, since it’s finally something from the Google on the level of o1 and Sonnet. So now we have 3 of them 🌚

You can try it out here https://aistudio.google.com/app/prompts/new_chat just make sure to choose 1206

Aider Code editing benchmark

To me, it's kinda the best benchmark out there right now in regard to coding capabilities

Totally aligned with my own experience of models usage

https://aider.chat/docs/leaderboards/

How sure you’re from 100% to 100% that it’s all gonna be fake and Gabe still never allows this magic to be used in any noscript?

Boeing 🤝 OpenAI