I strongly recommend that everyone try Jev themselves—it can boost your Codex operation speed by 10 times and save a ton of tokens!

Because Jev is perfect for judgment-based tasks, it can make accurate decisions in milliseconds, and the token price is extremely low. So it’s ideal for the following scenarios:

1、computer use Use Jev as the operation judgment layer to guide the model in quickly operating your computer https://x.com/gregpr07/status/2100411066966749359…

2、context compression Have Jev identify the key information in the context to achieve near-instant context compression https://x.com/tamarajtran/status/2100694549362553153…

3、model routing Automatically allocate the right model to handle tasks based on their complexity https://x.com/mdlahfir/status/2100314182201802811…

4、automatic review An automatic review model that’s even cheaper and more accurate than 5.6 luna, possibly the most suitable tool for Codex in safety approvals https://x.com/fazxes/status/2100300097695232164…

To use it, go to the official site https://typesafe.ai to apply—right now approvals are still fast, so everyone can come try it out and get a feel!

Home - TypeSafe AI

Source: https://typesafe.ai/

We took the opposite research direction

not chat

Reinforcement Learning from Human Feedback (RLHF) has led to LLMs that are optimized for human preferences. This has led to models that are superhuman at instruction following, and are what we now call “chat.” Yet RLHF creates inherent issues such as mode dropping, overconfidence, and lack of reliability. These flaws mean that LLMs require humans-in-the-loop.

a new model

We built a new class of models, System One Models, to be natively used by machines. We’re building with a new architecture, a new sampler, and a new training algorithm: Reinforcement Learning for Calibrated Decisions (RLCD).

Decisions, not strings

Typed outputs that software can act on.

calibrated confidence

Every decision includes an estimate of how confident the model is.

more like code

Reliable, fast, and type-safe.

[b.64]

ZmxvYXQgUV9yc3FydCggZmxvYXQgbnVtYmVyICkKewoJbG9uZyBpOwoJZmxvYXQgeDIsIHk7Cgljb25zdCBmbG9hdCB0aHJlZWhhbGZzID0gMS41RjsKCXgyID0gbnVtYmVyICogMC41RjsKCXkgID0gbnVtYmVyOwoJaSAgPSAqICggbG9uZyAqICkgJnk7CglpICA9IDB4NWYzNzU5ZGYgLSAoIGkgPj4gMSApOwoJeSAgPSAqICggZmxvYXQgKiApICZpOwoJeSAgPSB5ICogKCB0aHJlZWhhbGZzIC0gKCB4MiAqIHkgKiB5ICkgKTsKLy8JeSAgPSB5ICogKCB0aHJlZWhhbGZzIC0gKCB4MiAqIHkgKiB5ICkgKTsKCXJldHVybiB5Owp9

193.6x Faster, 444.6x Cheaper.

*based on workflows for System One tasks(proof)

TypeSafe AI

Cost $0.000081

Completed in 0.114s

LLMs

Cost $0.013880

Completed in 8.566s

Built for automation

Jev returns typed decisions with calibrated probabilities, so your software can account for uncertainty. Set the thresholds for when it acts autonomously and when it asks for review. Combine those decisions in code to build larger workflows, with control over how the intelligence is used.

Jev’s intelligence per dollar is literally off the charts.

Machine-Native Intelligence

LLMs produce words for people. Jev produces typed decisions and is more like code: reliable, fast, self-consistent, and type-safe.

Zero Hallucinations

Every Jev decision comes with a confidence estimate, so your software can act when confidence is high and escalate when it is not.

$42

Per Billion input tokens.

238x

Lower input price than Claude Fable 5.1

Come Build With Us

We give a FAQ

What are System One Models? What is Jev?

System One Models are a new class of AI model built for decisions inside software. Jev is TypeSafe’s first public System One Model, optimized for automation. Send Jev structured questions and get typed decisions with probabilities and confidence that your software can act on.

Is Jev just a smaller LLM?

How is this different from JSON mode or structured outputs?

How can Jev be so fast and inexpensive?

Can you make Jev even faster?

Are these prices temporary or subsidized?

What is Jev good at? Where does it struggle?

Can Jev still get things wrong?

How do I get started or ask a question?

[b.64]

iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAAAl0lEQVR42mNgGPyg5u9/e1xyCV9+/7XDJVn/G7eky5vfl+U5sMvZPPn9Ow6XobP//t2LS07k7+/XTjjkFM7+/V2HS2PG7787+HHIBXz4fVAcl6F///6dj8vQ6b9//1bHIWdw9/fvNbg0vvr9+wgPLsm/v39H4pKb///vX3lcNj75+70HR4AzOPz+ewdn/OOVlDiIR5K6AACSCULwD4UI6QAAAABJRU5ErkJggg==

VHlwZVNhZmUgQUkgSW50ZWxsaWdlbmNlIE5vdw==

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