Theodore

Building quickly

[Theodore @theoleecj]

@allenwlee Yes! Great news, @fcoury made a PR to support Macs and it has just been merged

[Theodore @theoleecj]

No waitlist, no funding, no hype, just Open Source Openjev - fast on-device (you can even run it in Chrome!) inference for type-safe decisions with probabilities!

https://t.co/S8G8iFTccl
https://t.co/LrM7jrIoSC

Built on Qwen 3 4B / 0.6B / 0.8B.

* no assocation with a Jev model

[Theodore @theoleecj]

@jkyamog Thanks so much for your pull request! Looking into it

[Theodore @theoleecj]

6 hours with Codex and Qwen 3.5 4B (or 0.8B) and you can get something eerily similar - and maybe open source??? https://t.co/S0EQnHkSGL 👀

Not 2 years in stealth!

It's Not Jev, it's something you can use today - in your browser, even!

[Diogo Almeida @CompleteSkeptic]

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

• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions

AFAICT the shortest path to AI-based economic revolution

[Theodore @theoleecj]

On the Jev waitlist?

You can use not Jev (but close enough) right here: https://t.co/S0EQnHkSGL

https://t.co/qLtRs0fHyD browser demo

[Theodore @theoleecj]

blueprint for DIY not-Jev

[Theodore @theoleecj]

@notkevinzhang https://t.co/S0EQnHkSGL
blueprint for DIY not-Jev

[Theodore @theoleecj]

@harshagundal Wow you beat me to it! I tried wrapping Qwen 4B directly and grabbing the logits and 0.8B for an in-browser demo https://t.co/S0EQnHkSGL

[Theodore @theoleecj]

https://t.co/S0EQnHkSGL
https://t.co/LrM7jrIoSC

[Theodore @theoleecj]

@adonis_singh Free and open source https://t.co/S0EQnHkSGL

[Theodore @theoleecj]

@jpschroeder I made an OSS (just wrap Qwen and grab the probailities)! https://t.co/S0EQnHkSGL

[Theodore @theoleecj]

@raunakdoesdev Yep, tried it and open sourced it https://t.co/S0EQnHkSGL

[Theodore @theoleecj]

@MichaelLee04 @hackgoofer Jev is cheap, but Openjev (no affiliation) is free~
https://t.co/pCIqL7NQvf

[chansung @algo_diver]

I think I could achieve similar results with @GoogleDeepMind Gemma3 270M.

[Theodore @theoleecj]

@notkevinzhang You can skip a waitlist with https://t.co/S0EQnHkSGL or even try something out in your browser with https://t.co/LrM7jrIoSC It's not Jev, it's just something you can try today

[Theodore @theoleecj]

It's Not Jev, and you can use it today, without waitlist - in your browser, or cloned to your PC.

[Theodore @theoleecj]

@CompleteSkeptic If you dont wanna sit on a waitlist, try out https://t.co/LrM7jrIoSC (no affiliation, and no waitlist too!)

[Theodore @theoleecj]

We also compare this approach of using logits + Qwen 3.5 4B against Qwen reranker and find satisfactory performance.

[Theodore @theoleecj]

Its accuracy with Qwen 3.5 4B almost matches another currently-closed-source, waitlisted (at time of writing) model on some public evaluations like https://t.co/NAdoJAcEdC and https://t.co/8zQXeuJ7Py, obtaining 0.845 against the closed model's 0.883.