Diogo Almeida

bluesky's tests showed us winning on accuracy, cost, and speed of not just a frontier model, but a FINE TUNED frontier model that's just insane

hailey@hailey.at

okay i’m very impressed with @typesafeai.bsky.social. it’s a bit late and it’s time to have some beers in mexico but i’m getting some extremely promising results on real world classification tasks that historically would have needed either large models or finetunes on expensive gpus

Message from our CEO: After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev

View full thread

typesafe.aiIntroducing System One Models and Jev - TypeSafe AI BlogTypeSafe AI is a new AI lab operating in stealth

Extraordinary claims require extraordinary evidence so check out our release blog for more technical info: typesafe.ai/blog/introdu... Join our waitlist for early access: typesafe.ai

discord.ggJoin the TypeSafe AI Discord Server!Check out the TypeSafe AI community on Discord - hang out with 219 other members and enjoy free voice and text chat.

Have technical chats and meme with us on Discord (rumors are good memers skip the line): discord.gg/WUujKYBp8s We are so pumped this is in the hands of developers now, and we are just getting started!!!

View full thread

We love how this doomo doomonstrates real-time intelligence and what can be doone with code + AI! ~10 calls/sec = ~$7/hour

Game: race from one Wikipedia page to another using only links Challenge: choosing between hundreds to thousands of links Shows not just intelligence-per-second, but also the compounding benefits of not hallucinating with high-cardinality choices

View full thread

The gains aren’t free: Jev can't generate text Comparing Jev vs LLMs side-by-side makes the trade-off clear Fun fact: replacing sequential computation with parallel is the same way Transformers leapfrogged RNNs

We believe that the future is code + AI, so made workflow evals to reflect that Jev costs: $42 / BILLION input tokens ($0.042 / MTok) and output tokens are free (forever - they’re too cheap to meter with our new architecture) Jev is named after Jevons paradox and off the intelligence per $ charts