Architecture Decisions

Jev vs LLMs: Cost, Speed and Where Each Fits

What Jev is, what classification costs on Jev versus Claude, GPT and Gemini at list price, where it falls short, and the alternatives worth testing.

Manmeet SinghManmeet SinghFounder, AIshar Labs · Ex-Apple, Ex-Instacart, Ex-Adobe
9 min readOctober 2026

A lot of LLM calls inside software don't write anything. They read some text and pick one of a few options. Is this a refund request or a bug report? Does this action need approval? Is this message spam? You can send those through Claude, GPT or Gemini and it works. It's also a lot of model for the job.

Jev is built for exactly that job. It's a classification model from a startup called TypeSafe, launched in September 2026. You give it text and a question, and it gives back one of the answers you defined, with a probability and no prose. At list price, a million classifications on Jev cost about $24. The same million costs about $68 on the cheapest small LLMs when they answer with only a label, about $1,350 on mid-tier models doing the same, and over $50,000 on flagship models that write a long answer first.

That gap is real, but its size depends on what you compare against, and nearly all the evidence for Jev so far comes from TypeSafe. Here is what's known, what isn't, and how to test it on your own data.

What is Jev?

Jev is what TypeSafe calls a decision model, the first of its "System One" models. It doesn't generate text. It returns a typed answer: yes or no with a probability, a pick from a list with a probability for each option, or a score from 0 to 10 worked out from those probabilities, as The Batch describes it.

You send a block of text and one or more questions. The questions run in parallel, you pay for the text once plus the question tokens, and output is free. The OpenRouter listing shows $0.042 per million input tokens and $0 for output.

Two things matter if you plan to depend on it. TypeSafe says it trained Jev so its probabilities mean something: an answer it gives 20 percent should be right about 20 percent of the time. And TypeSafe says Jev isn't an LLM and doesn't generate token by token, although it's built on a transformer. Most of the architecture isn't public, and the sources I read disagree on the context limit, so check with TypeSafe before you design around a number.

How is Jev different from an LLM?

An LLM writes an answer. Jev picks one. In practice that means three differences.

How we're using it

We've started replacing LLM calls with Jev at decision points in the systems we build, the places where a system has to classify some text or choose between a few options before it does anything else. Those calls don't need a model that can write, and I'd rather not pay for one or wait on one.

I don't have a clean before-and-after number from client work that I can publish, so the cost figures below are list-price math, not our invoices. Use them to size the opportunity, not as a promise.

The pattern I reach for is a classifier at the front door. It decides, and only the uncertain or genuinely hard cases go on to an LLM or a person.

How much cheaper is Jev than Claude, GPT and Gemini?

It depends on what you're comparing against. I priced one million classifications of a 500-token text at each provider's list price, read from their official pricing pages on October 20, 2026.

What one million classifications cost at list price
LLM answers with only a label (about 5 tokens)LLM writes a 900-token answer firstJev (output is free)
Cost of one million classifications by model, log scaleJev costs about $24. Small language models that answer with a label cost $68 and up, and flagship models that write long answers cost up to $51,500. The table below lists every value.$10$100$1K$10K$100KDollars per 1 million classifications (log scale)Jev 1.13: $24 per million classificationsJev 1.13$24Claude Haiku 5.5: $68 with a label-only answer, $515 with a 900-token answerClaude Haiku 5.5$68 – $515GPT-6 Luna: $68 with a label-only answer, $515 with a 900-token answerGPT-6 Luna$68 – $515Gemini 3.5 Flash-Lite: $208 with a label-only answer, $2,445 with a 900-token answerGemini 3.5 Flash-Lite$208 – $2,445Gemini 3.8 Flash: $506 with a label-only answer, $3,863 with a 900-token answerGemini 3.8 Flash$506 – $3,863Claude Sonnet 5.5: $1,350 with a label-only answer, $10,300 with a 900-token answerClaude Sonnet 5.5$1,350 – $10,300GPT-5.6 Terra: $1,360 with a label-only answer, $12,100 with a 900-token answerGPT-5.6 Terra$1,360 – $12,100Claude Fable 5.1: $6,750 with a label-only answer, $51,500 with a 900-token answerClaude Fable 5.1$6,750 – $51,500GPT-6 Astra: $6,750 with a label-only answer, $51,500 with a 900-token answerGPT-6 Astra$6,750 – $51,500

Illustrative workload, not a measurement: a 500-token text. LLMs get about 650 input tokens (text plus instructions). Jev gets 580 (text plus four short questions) and bills no output. List prices per 1M tokens read from each provider’s official pricing page on October 20, 2026. Gemini 3.8 Flash doubles on January 1, 2027.

Cost per million classifications by model, with the multiple of Jev’s cost
ModelInput $/1M tokensOutput $/1M tokensLabel-only answer900-token answerx Jev (label)x Jev (written)
Jev 1.13$0.042$0$24$241x1x
Claude Haiku 5.5$0.1$0.5$68$5152.8x21x
GPT-6 Luna$0.1$0.5$68$5152.8x21x
Gemini 3.5 Flash-Lite$0.3$2.5$208$2,4458.5x100x
Gemini 3.8 Flash$0.75$3.75$506$3,86321x159x
Claude Sonnet 5.5$2$10$1,350$10,30055x423x
GPT-5.6 Terra$2$12$1,360$12,10056x497x
Claude Fable 5.1$10$50$6,750$51,500277x2,114x
GPT-6 Astra$10$50$6,750$51,500277x2,114x

Here's how I read it.

That last band is where TypeSafe's "up to 400x cheaper" headline lives. It's accurate there. It isn't what you'll save if you're already running a small model with a one-word answer.

Put in dollars, ten million decisions a month would cost roughly $240 on Jev. On Claude Haiku 5.5 with a label-only answer it's about $675. On Gemini 3.8 Flash it's about $5,060. On Claude Sonnet 5.5 it's about $13,500 with a label-only answer and about $103,000 if the model writes 900 tokens each time.

So the saving is mostly a story about avoided waste. If a routing decision goes to a frontier model that reasons at length, move it. If it's already on a small model, choose on speed and calibration, not price. This math also uses list prices. Caching and batch discounts will pull the LLM numbers down, and Jev's bill grows with the number of questions you ask per text.

Is Jev faster?

TypeSafe reports 70 to 500 milliseconds per request. Averaged across its own four internal workflows, it reports 0.4 seconds per case for Jev against 10 seconds for GPT-5.6 Terra and 38 seconds for Opus 5, as Arize relays them.

Average time per case, as reported by TypeSafe
Average seconds per case across TypeSafe’s four internal workflowsJev 0.4 seconds, GPT-5.6 Terra 10 seconds, Opus 5 38 seconds. Reported by TypeSafe and not independently verified.Jev: 0.4 s per case (TypeSafe-reported)Jev0.4 sGPT-5.6 Terra: 10 s per case (TypeSafe-reported)GPT-5.6 Terra10 sOpus 5: 38 s per case (TypeSafe-reported)Opus 538 sSeconds per case, linear scale from zero

Averages across TypeSafe’s four internal workflows, reported by TypeSafe and relayed by Arize in September 2026. Arize had not run its own benchmark at the time. The larger models were run with reasoning, which accounts for much of the gap.

Two cautions. TypeSafe measured this itself. And the bigger models were run with reasoning, which accounts for much of the gap. A small LLM answering with one word will be much closer. Another test Arize cites clocked 777 judgments in under 0.7 seconds, which shows how quickly Jev handles a batch. Time the 95th percentile on your own traffic before you believe any of these numbers.

Is Jev as accurate as an LLM?

Nobody outside TypeSafe knows yet. This is what exists.

The spam test is the interesting one. A model that matches a trained classifier with no labeled data could save you weeks. It's still one test.

The calibration result is the part I'd pay attention to. In that test, emails scored under 0.1 were 0.1 percent spam, and emails scored 0.9 or above were 99.9 percent spam. The 0.5 to 0.6 band was 38 percent spam. Sending the 4.6 percent of emails scored between 0.3 and 0.7 to a person left the rest at 99.5 percent accuracy. That's the real prize: a probability you can set a threshold on.

What are the alternatives to Jev?

A few options do similar work. Most of the open ones are days or weeks old, and their claims haven't been checked.

OptionWhat it isWhere it's strongWatch out forSource
JevHosted decision model that returns typed answers with probabilitiesCost per decision, speed, no training dataEarly access, no explanations, mostly vendor-reported resultsThe Batch, Arize
Small LLM with a constrained outputClaude Haiku 5.5, GPT-6 Luna or Gemini 3.5 Flash-Lite told to answer with a labelAlready in your stack, flexible, can explain when askedAbout 3x to 9x Jev's cost on our example even at label-only, calibration variesProvider pricing pages
LayaClassification model published on GitHub, with a focus on multilingual supportLanguages beyond EnglishIts claim to beat Jev is unverifiedThe Batch
KevModel on GitHub built on Qwen 3.5 in three sizes, trying to reconstruct Jev's architecturePublished code you can inspectUnverified, and a reconstruction, not a copyThe Batch
Bespoke NimbleA Qwen-3.5-9B model fine-tuned as a classifier, published on GitHubPublished code you can inspectSame: unverifiedThe Batch
Classic classifier (TF-IDF with logistic regression)A trained model on labeled examplesMatched Jev's accuracy on one spam test and costs almost nothing to runNeeded about 14,800 labeled examples, struggles when categories changeArize
Rules and plain codeHand-written logicPredictable and freeBreaks on messy languageMindStudio

When should you use Jev, an LLM, a classic classifier or plain code?

The safety gate is the one I'd watch. An agent that proposes dozens of actions can have each one checked cheaply, which is hard to afford when every check is a frontier-model call. It's one reason decision models matter for agent systems.

How to test it on your own data in a day

1. Pull 300 to 500 real examples and have a person label them. 2. Write each decision as several short, separate questions instead of one broad one. The Batch recommends this. 3. Run Jev, a small LLM told to answer with a label, and your current setup on the same examples. 4. Compare accuracy. Then check calibration: of the cases Jev scores 0.9 or higher, are at least 90 percent right? 5. Time the 95th percentile on real traffic and price it with your actual token counts. 6. Pick a threshold, send the uncertain band to an LLM or a person, and cost that too. 7. Run it again every month. This market is moving fast.

Quick answers

Is Jev an LLM?

No, according to TypeSafe. It's transformer-based, but it returns a choice instead of generating text one token at a time.

Does Jev hallucinate?

TypeSafe says it can't. Arize calls that an overreach. Jev can't answer outside the options you give it, but an answer inside them can still be wrong.

Can Jev replace Claude, GPT or Gemini?

For deciding, often. For writing, explaining or reasoning, no.

Who makes Jev?

TypeSafe, a startup founded by OpenAI alumnus Diogo Almeida, according to The Batch. It launched in early access in September 2026 and is also listed on OpenRouter.

How much does Jev cost?

$0.042 per million input tokens and nothing for output, per the OpenRouter listing.

If you're paying LLM prices for decisions

If you suspect your system is paying for a model that writes where a model that chooses would do, that's a good first look for a two-week Architecture Sprint. Our LLM cost optimization work starts with exactly this question: which calls need a big model, and which don't.

Sources

Checked October 20, 2026. Prices change, so confirm before you budget.

Want help with your AI stack?

If this post matches problems you're seeing, we can map the fastest path from architecture decisions to production outcomes.

Book a call
Architecture DecisionsAI EngineeringProduction ML
Manmeet Singh

Manmeet Singh

Founder & CEO, AIshar Labs · Ex-Apple, Ex-Instacart · 15 AI Patents Filed

Built ML systems at Apple (Search: Maps, Safari, Spotlight) and Instacart (Search, Recommendations, Ranking). Writes about production AI tradeoffs and system design.

Follow on LinkedIn →

More from AIshar Labs