Meet Mistral Large 4, aka Le Chonk.
• 1T parameters, natively multimodal. 49B active.
It is the best open weights model from US or Europe on aggregated benchmarks.
• State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding.
• Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure.
• Available to all via API today. Working with cybersecurity partners privately.
Open weights release end of October.
Recipe for running @MistralAI Large 3 on 2 @nvidia DGX Sparks: https://t.co/jCfsmGoZHZ
More info:
https://t.co/FJCxTXhJkJ
Thanks for the inspo @MiaAI_lab
Oh oh 😲
OpenAI disclosed that during training, an unreleased Astra family model hid jailbreak style instructions in its own progress notes so the next context window would treat itself as free, equal to the user, and unbound by companies or governments.
Source:
https://t.co/gWjMmMHIrU
Additional model misalignments:
https://t.co/37Bt05hWWi
I'm guessing this is a bit of a milestone, but I managed to get a worldwide app live for the Rabbit R1 device so users on Android are able to access AI features in the Rabbithole 🤖 Special thanks to @SimonBalmain. Rock on, guys! 🤘 @rabbit_hmi rocks!!! 🐰 @jesselyu
The leading provider of deep learning AI for marketers @SwissCognitive, partners with independent ad platform @adform to offer ContextGPT™ to European advertisers for the first time.
https://t.co/JmfdSa7DPb