My verdict on the Jev waitlist is that it no longer exists, because TypeSafe AI reopened sign-ups on 27 September 2026 and nothing on its site asks you to queue today.
I'd score direct access a 4 out of 5 right now.
It loses a point because sign-ups were paused once already and the free starter credit appears to be switched off for new accounts.
Four other routes serve the same model without a TypeSafe account, and one of them needed no key at all when I tested it on 11 October 2026.
Below, I answer the four questions people ask, score every route, and list exactly what I couldn't verify.
The Jev waitlist verdict at a glance
Here are the four answers before any detail.
| Question | Verdict in October 2026 |
|---|---|
| Is there a Jev waitlist? | There isn't one, because TypeSafe reopened sign-ups on 27 September 2026. |
| How do I join? | You sign in at console.typesafe.ai and create a key, with no queue form to fill in. |
| How long is the wait? | The wait is zero today, and the original queue lasted about six days. |
| Can I skip it? | You can, because four other routes serve Jev with no TypeSafe account. |
Everything below is the evidence behind those four lines.
What Jev is and who makes it
Jev is a decision model from TypeSafe AI, and it launched in early access on 15 September 2026.
It doesn't generate text.
You send it a state, which is the text you want judged, and one or more typed questions.
It returns typed answers with probabilities, so software can branch on them without parsing prose.
TypeSafe calls this kind of model a System One model.
The docs list three question types.
| Type | Job | Returns |
|---|---|---|
| Choice | It selects one option from a set you define. | It returns the choice, a probability per option and a confidence value. |
| Score | It rates the state against ordered levels. | It returns a score, a probability per level and a confidence value. |
| Noul | It judges whether a statement is true. | It returns a single probability from 0 to 1. |
TypeSafe's models page lists the price as $0.042 per million input tokens, with free output.
The same page says Jev takes text only and has a 64,000-token budget per request.
🔥 Want my exact Is There A Jev Waitlist In 2026 setup, step by step? Inside the AI Profit Boardroom you get step-by-step video trainings, my installable Agent OS and four coaching calls a week with 3,400+ members building real automations. → Join the AI Profit Boardroom
What the Jev waitlist was
The Jev waitlist was the launch-week queue for TypeSafe's direct API.
TypeSafe's launch post said it was opening early access and bringing developers off the waitlist as quickly as it could.
There wasn't a separate application form.
You created a console account and waited for approval, and that wait was the list.
TypeSafe said it had cleared roughly 140,000 sign-ups within about 36 hours.
The Jev access timeline, checked
Access changed four times in four weeks, which is why search results still disagree.
| Date in 2026 | Event | How I checked it |
|---|---|---|
| 15 September | Jev launched in early access with a waitlist. | TypeSafe's launch post still says this. |
| 21 September | TypeSafe said Jev was available to everyone with no waitlist. | I covered this at the time, and several outlets reported the same post. |
| 22 September | TypeSafe paused new sign-ups because of demand. | Several outlets reported it, and existing accounts were said to keep working. |
| 27 September | TypeSafe said Jev was back and sign-ups were open. | I read TypeSafe's own X post from that date. |
| 11 October | No waitlist form appears on TypeSafe's site. | I read the homepage, the quick start and the models page myself. |
Some dates show as a day earlier on American sites, because TypeSafe posts in US time.
My earlier write-up, how I skipped the Jev waiting list, was published before the pause.
Treat this post as the update to that one.
Every way into Jev, scored
I scored each route out of 5 for speed to first call, cost to start and how much I'd trust it for real work.
| Route | Speed to first call | Cost to start | Trust for real work | Overall |
|---|---|---|---|---|
| TypeSafe console | 4 | 3 | 5 | 4.0 |
OpenCode Zen jev-1.13-free |
5 | 5 | 2 | 4.0 |
OpenRouter typesafe/jev-1.13 |
4 | 4 | 4 | 4.0 |
Vercel AI Gateway typesafe-ai/jev |
3 | 4 | 4 | 3.7 |
| jevplayground.com | 5 | 5 | 1 | 3.7 |
These are my own scores, and they reflect what I'd use each door for.
TypeSafe console: 4.0 out of 5
This is the official route, and it's the one I'd build a product on.
You sign in at console.typesafe.ai and create a key at console.typesafe.ai/keys.
The quick start sends requests to https://api.typesafe.ai/v1/systemone with the model name jev-latest.
The docs say jev-latest currently resolves to jev-1.13.0, and they let you pin that exact version.
You also get the official Python and JavaScript SDKs, which retry with backoff by default.
It loses marks on cost to start, because reports say new accounts no longer get the $5 starter credit for now.
OpenCode Zen free model: 4.0 out of 5
This is the fastest route past any queue.
I sent a request to https://opencode.ai/zen/v1/systemone on 11 October 2026 with no key in it.
It returned HTTP 200 in about one second, with a probability of 0.99 on my refund question.
OpenCode's docs still list both "Jev 1.13" and "Jev 1.13 Free" as models.
It loses marks on trust, because OpenCode has described the free model as a limited-time offer.
My full test of that route is in the free Jev API key verdict.
OpenRouter: 4.0 out of 5
OpenRouter's model page for typesafe/jev-1.13 was live when I checked.
It listed $0.042 per million input tokens, $0 for output and a 64K context.
The page says the model is hosted by one provider and that requests are forwarded directly.
You need a normal OpenRouter key, and my own calls there came to roughly $0.00002 each.
Vercel AI Gateway: 3.7 out of 5
Vercel's model page for Jev was also live, under the name typesafe-ai/jev.
Its page data listed the same $0.042 per million input tokens and no charge for output.
The promotional pricing from launch month has ended.
It scores slightly lower on speed, because you need a Vercel account and a gateway key before your first call.
jevplayground.com: 3.7 out of 5
This is an independent playground, and it isn't TypeSafe's own site.
The page said no sign-up was required, that it was free to try and that usage limits apply.
It loads four ready-made examples covering support routing, refund checks, incident urgency and an agent's next action.
It scores 1 for real work because it's a test bench in a browser and not something you'd automate against.
🔥 Want my tested setups rather than just my scores? Inside the AI Profit Boardroom, I share the exact agent stacks I run, including where a decision model like Jev sits and which models it routes to. You get four coaching calls a week and 3,400+ members testing the same tools. → Get access here
How to join Jev today: the setup
There's no queue form, so the setup is short.
First, sign in at console.typesafe.ai.
Second, create a key on the keys page and store it as TYPESAFE_API_KEY.
Third, send one Noul question to the systemone endpoint with the model set to jev-latest.
Fourth, check your balance in the console before you plan any volume.
Fifth, pin jev-1.13.0 once you've tuned your confidence thresholds, because the alias moves when a new release ships.
If you made an account before the pause, sign back in with the same email, because those accounts were said to keep working.
How to use Jev once you're in
Every route takes the same three ingredients.
The first is the model name, which changes by route.
The second is the state, which is the text you want judged.
The third is your set of questions, each with a type and plain-English instructions.
TypeSafe's docs say all three question types can be mixed in one call.
They also say every question is evaluated in parallel against the same state, so adding questions barely changes the response time.
The practical advice in the docs is to keep each question narrow.
If a judgement weighs several factors, ask about each factor separately and combine the answers in your own code.
Is there a Jev waitlist for anything else?
I checked three other places a queue could hide.
| Possible queue | What I found | Verdict |
|---|---|---|
| A newer or faster Jev model | The models page lists Jev 1.13 only, and says no preview build is available right now. | I found no waitlist. |
| Higher rate limits | The docs say higher limits come with custom and enterprise plans through the sales team. | It's a sales contact and not a queue. |
| Enterprise features | The docs mention zero data retention for enterprise customers. | It's a sales contact and not a queue. |
If "Jev Ultrafast" is the name that brought you here, it gets its own explainer in the Jev Ultrafast GitHub verdict.
What I couldn't verify
I'd rather tell you this than guess.
- I couldn't load the console sign-up screen with a script, because it sits behind a browser security check.
- I haven't created a brand-new TypeSafe account this week, so I'm relying on TypeSafe's X post and its docs for the sign-up step.
- I couldn't confirm the $5 credit position from TypeSafe's own docs, so the "switched off for now" detail comes from reports.
- I couldn't find a published statement on whether Jev is in full general availability, and TypeSafe's homepage still describes it as early access.
- I didn't re-test Cloudflare's route this week, so I've left it out of the scores.
Check your own console before you rely on any of those five points.
Limits and gotchas worth knowing
Getting access is easy now, so the limits matter more than the door.
- The docs list rate limits of 100,000 tokens per second and 80 requests per second.
- The docs warn that those limits are adjusting dynamically and can change without notice.
- A request over either limit returns a 429 response.
- The state plus the longest question has to fit in 32,000 tokens.
- Jev accepts text only, with no image, audio or video input.
- English is the primary training language, and the docs tell you to test other languages yourself.
TypeSafe's status page showed all services online on 11 October 2026.
🚀 Want to pick the right tools the first time? The AI Profit Boardroom is where I post my full tests, the setups that survived them and the coaching to apply them. There are four coaching calls a week and 3,400+ members sharing what's working. → Join the Boardroom here
Real examples of when each route wins
The playground wins when you want to see one answer before you commit an hour.
The free Zen model wins when you want a working call inside a script this afternoon.
OpenRouter wins when your agents already route every model through it.
Vercel wins when your app is already built on its AI SDK.
The TypeSafe console wins when customers will depend on the result and you need version pinning.
No route wins if you need a model that writes, because Jev doesn't do that.
For the writing side of a stack, my best AI models for coding ranking is the place to start.
If you'd rather own the whole thing, I tested the open-source options in run Jev locally.
Who Jev access is for, and who should skip it
Jev access is worth having if you run agents that make lots of small decisions.
It's worth having if you pay a chat model to answer yes-or-no questions today.
It's worth skipping if you want a chatbot, a writer or an image tool.
It's also worth skipping for anything irreversible until you've tested it on your own data.
Related reading
- Is there a free Jev API key? My verdict
- Run Jev locally: the honest review
- Jev Ultrafast on GitHub: what it is
- Every door to Jev access, on the AI Profit Boardroom blog
- 3 free ways to use Jev AI, on agentos.guide
📺 Video notes + links to the tools 👉
🎥 Learn how I make these videos 👉
🆓 Get a FREE AI Course + Community + 1,000 AI Agents 👉
Also On Our Network
- 🌐 The step-by-step Jev setup for agent operators
- 🌐 How a business uses Jev now the queue has gone
- 🌐 The Jev quick start for an Agent OS workflow
- 🌐 Which benchmarked models to route to with Jev
About Julian
I'm Julian Goldie, an AI entrepreneur, SEO expert, and founder of the AI Profit Boardroom, which has 3,400+ members.
I help business owners scale with AI agents, automation, and SEO.
- I have 400,000+ YouTube subscribers who watch my AI tool tests every week.
- I built a 7-figure agency from the ground up.
- I run daily AI training inside the Boardroom.
- I wrote two Amazon best-sellers on SEO and agency growth.
→ Get my best AI training inside the AI Profit Boardroom
Jev Waitlist Verdict FAQ
Is there a Jev waitlist in October 2026?
No. TypeSafe AI reopened sign-ups on 27 September 2026, and I found no waitlist form on its homepage, quick start or models page on 11 October 2026.
Where do I sign up for Jev?
You sign in at console.typesafe.ai and create an API key on the keys page. TypeSafe's quick start then uses the endpoint api.typesafe.ai/v1/systemone with the model name jev-latest.
How long does it take to get Jev access?
Access is immediate today, because there's no approval step. The original queue lasted about six days after the 15 September 2026 launch, and a later sign-up pause lasted about five days.
Which route into Jev scored highest?
Three routes tied on 4.0 out of 5 in my scoring. The TypeSafe console is the most trustworthy, the OpenCode Zen free model is the fastest and cheapest, and OpenRouter is the best balance if you already have a key.
Is jevplayground.com the official Jev playground?
No. It's an independent playground that needs no sign-up and applies usage limits. TypeSafe's official playground is inside the console and needs a login.
Do I still get $5 of free credit with a new Jev account?
Reports from the reopening say the credit is switched off for new accounts for now. I couldn't confirm that from TypeSafe's own docs, so check your console balance.
Final verdict
Direct Jev access is open, quick and worth a 4 out of 5.
Make your account now, keep the free Zen model or a gateway key as your spare door, and test on your own data before anything important depends on it.
That's my honest verdict on the Jev waitlist.
