∵ ⩆
⩆ ∵
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
Watch the real video
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.
Jev.Cost
$42
Per Billion input tokens.
238x
Lower input price than Claude Fable 5.1
Come Build With Us
∵ ⩆
⩆ ∵
TypeSafeAI Blog
⩆ ∵
No items
Thoughts
The Bitterest Lesson
Read More
TL;DR: Compute drives progress in AI, but what good is progress if you are not doing the right task!
Thoughts
AI: too good to be true, too bad to be useful | TypeSafe AI
Read More
RLHF-trained language models please humans and assist rather than make reliable autonomous decisions. What comes next?
∵ ⩆
⩆ ∵
∵ ⩆
⩆ ∵
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?
Is Jev deterministic?
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==