Sam Sorenson @samtypesafe
Joined September 2026-
Tweets41
-
Followers2
-
Following2
-
Likes101
Yes, classifiers existed. BERT, DeBERTa, GLiNER, fine-tuned heads. Most of them were specialized. Pick the labels, train, ship. Changing the rules at runtime meant another training run. What we did not have was a general-purpose classifier at Terra / Sonnet 5 level intelligence. You could fake it by asking an LLM for JSON, but that was slow, expensive, and poorly calibrated. Jev is that missing piece. Labels defined in code, typed answer plus confidence, no extra LLM call. Classification is not new. What is new is a general-purpose classifier this intelligent, this fast, and this cheap.
I'm genuinely surprised people didn't know Classification models before!
if you're an investor and suggest raising prices, you will have a VERY BAD time
Investors just asked @CompleteSkeptic if we would raise prices and boy did they regret that We're profitable at current prices. We're gonna make it cheaper. We're here for the builders. Jev: on
Jev by @typesafeai is now on the Venice API, in beta. It answers, it doesn't write. Send it your app's state and typed questions. Get back a typed answer your code can branch on: a probability, a chosen option, or a score on your rubric. No JSON to coax out of a chat model.
Jev was adopted faster than any other model in AI Gateway history. In the first day, @typesafeai reached ~13% of teams, 2x the GPT-5.6 family and 6x Fable 5.1.
Jev from @typesafeai is now live on @CloudflareDev AI Gateway. Try the first System One model — send state and typed questions; get structured answers your code can use directly. developers.cloudflare.com/ai/models/type…
100% of the data is synthetic people are often skeptical of synthetic data. it just proves how carefully constructed synthetic data can be incredibly powerful
@badlogicgames you might be the first person talking about the data over the architecture! 🥲 we consider ourselves a data research lab! the vast vast vast majority of research was on making data that is truly general (ala a cognitive core) and 100% of our data is synthetic (but not the type
Behind the scenes of our launch video… chief of staff @justKDeng keeping our CEO @CompleteSkeptic on track
how are more people not talking about "The Bitterest Lesson" from @typesafeai this is one of the most insightful posts i've read recently
update: jev is live on @Cloudflare ai gateway 🪅
do i know anyone at @typesafeai? would love to get jev on @Cloudflare ai gateway
so sorry it's going down! demand is beyond our wildest dreams 😭 we really want this up and available for everyone
update: jev is live on @Cloudflare ai gateway 🪅
people think we can’t afford to serve jev sustainably at current prices i’m telling you as someone close to the magic that we 100% can and we’re just getting started prepare for the floor to collapse on intelligence prices
just got access to JEV and replaced my prefrontal cortex with it, I’m finally free from the burden of choice thank you @typesafeai
one of the most interesting bit from the jev release
Jev's Secret Sauce
I made Jev play Flappy Bird against 3 other models. Who do you think won?
down with public evals up with evals for your REAL WORK 👏
We benchmarked fx auto mode (safety) classifier with @typesafeai's Jev. tl;dr: ~5-18x faster and more accurate than 𝚐𝚙𝚝-𝟻.𝟼-𝚕𝚞𝚗𝚊, our current top choice
Incredibly exciting that there are entire universes of AI innovation that still exist that weren’t even on most of our radars. Being able to process information insanely quickly, at crazy low costs, with high levels of capability is huge for a wide number of enterprise tasks. Data classification tasks, routing decisions inside of a workflow, decision making when handed a particular domain problem, quick judgment calls about safety or security, and more all are the gates in a large number of processes. This model and approach could be quite cool in agentic workflows in the enterprise.
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
the grifters are dropping 1b fine tunes with constrained decoding and saying it’s the same as Jev. Way to out yourself
They were building in stealth for 2 years, I was building in stealth for 2 hours… Happy to open source Qwen-2.5-1B-RLCD, 5x faster on-device inference for JSON workloads that need to be type-safe. ⚡️Demo below on a M4 MacBook⚡️ every LLM has the ability to efficiently batch
BABAA♥️ @Baba2478011
606 Followers 4K Following There is only one way to avoid criticism: do nothing, say nothing, and be nothing✌🏾💙
Valerie Chocklett @val_typesafe
2 Followers 19 FollowingYou might like
