Reasoning with latency-optimized models is quite a UX game changer. Super proud of what the team has accomplished with this Magistral release! mistral.ai/news/magistral

My team (reasoning) is hiring: jobs.lever.co/mistral?team=R… If you're interested in building/improving/using this kind of RL pipeline and do cool stuff, join us!

Very important: if you’re evaluating Magistral, plz use the system prompt here: huggingface.co/mistralai/Magi… Or better still, put the system prompt in the user message.

We purposely made it great at optimal transport as you may have guessed !

"What is a rabbit?" "A rabbit is what you have in the absence of non-rabbits" -- QA session at Tim Gowers's talk in Big Proof, paraphrased

How to train a reasoning model, specifically Magistral from Mistral AI? Here is a quick video walkthrough of the paper.

Call me Shinjishmael

i'm not sure this duo lingo course is that popular but it filled my need

Masterminding superintelligence. Credits to Jason Rute

DSP just got an upgrade!

Happy to introduce Kimina-Prover-72B ! Reaching 92.2% on miniF2F using Test time RL. It can solve IMO problems using more than 500 lines of Lean 4 code ! Check our blog post here: huggingface.co/blog/AI-MO/kim… And play with our demo ! demo.projectnumina.ai

> be me > using a new browser > want to use bionic reading that highlights word prefixes for faster reading > don't remember its name > search "adhd reading chrome extension" > got it as the first result

Bastille day is a French holiday. No meetings throughout the entire day and got loads of IC work done. Thank you revolutionary insurgents 236 years ago for making me productive.

Introducing the world's best (and open) speech recognition models!

Introducing powerful new features in Le Chat, making it more capable and more fun!

Pretty happy with how my predictions are holding up. 5/6 was the gold medal threshold this year. OAI's "experimental reasoning LLM" got that exactly, failing only to solve the one hard combinatorics problem, P6. My advice remains: look beyond the medal. Brief thread. 1/

I think machine learning paper titles will benefit a lot from using existential and universal quantifiers