Decision model pricing calculator: what Jev, Clef, Decider and Luna Decisions cost for your workload
Enter how many evaluations you run a month, how long the text is and how many questions each call asks. The table estimates input tokens per call for all five decision models, the monthly bill at each vendor’s list price, and the cost on JevStation. The token counts come from our own calls on 2026-10-07, because list prices alone mislead: Perplexity Decider has the lowest price per token but bills the text once per question.
JevTypeSafe · $0.042 / 100만 입력 토큰
- 직접 이용 시 월 비용
- $2.66
- JevStation 월 비용
- $137.10
- 입력 토큰 / 호출
- 634
- 크레딧 / 호출
- 1
Clef-flashCloudflare · $0.09 / 100만 입력 토큰
- 직접 이용 시 월 비용
- $5.53
- JevStation 월 비용
- $137.10
- 입력 토큰 / 호출
- 614
- 크레딧 / 호출
- 1
ClefCloudflare · $0.24 / 100만 입력 토큰
- 직접 이용 시 월 비용
- $14.74
- JevStation 월 비용
- $270.30
- 입력 토큰 / 호출
- 614
- 크레딧 / 호출
- 2
DeciderPerplexity · $0.04 / 100만 입력 토큰
- 직접 이용 시 월 비용
- $3.94
- JevStation 월 비용
- $270.30
- 입력 토큰 / 호출
- 986
- 크레딧 / 호출
- 2
Luna DecisionsOpenAI · $0.10 / 100만 입력 토큰
- 직접 이용 시 월 비용
- $6.84
- JevStation 월 비용
- $137.10
- 입력 토큰 / 호출
- 684
- 크레딧 / 호출
- 1
Decision-1Microsoft · $0.042 / 100만 입력 토큰
- 직접 이용 시 월 비용
- $1.47최저가
- JevStation 월 비용
- $137.10
- 입력 토큰 / 호출
- 350
- 크레딧 / 호출
- 1
측정된 호출에 맞춰 만든 토큰 모델로 추정한 값입니다. 다섯 모델 모두 출력 토큰은 무료입니다. 직접 이용 가격에는 게이트웨이 수수료가 포함되지 않으며, JevStation 열은 가입 크레딧 50개를 쓴 뒤 일회성 크레딧 팩을 가장 저렴하게 조합한 금액입니다.
How the estimate works
Each model’s row uses a token formula fitted to nine or more real calls per model on 2026-10-07 (three text lengths × one, three and five questions), then multiplies by the list price per million input tokens.
| Model | List price, input | Input tokens per call, our fit |
|---|---|---|
| Jev | $0.042 / M | 260 + 0.163 × characters + 43 × questions |
| Clef-flash | $0.09 / M | 29 + 0.168 × characters + 111 × questions ¹ |
| Clef | $0.24 / M | 29 + 0.168 × characters + 111 × questions ¹ |
| Perplexity Decider | $0.04 / M | questions × (84 + 0.163 × characters) |
| OpenAI Luna Decisions | $0.10 / M | 22 + 0.163 × characters + 139 × questions |
| Microsoft-Decision-1 | $0.042 / M | 0.163 × characters + 35 × questions ² |
¹ On Cloudflare Workers AI on 2026-10-04 the text part stopped counting at about 2,048 tokens because the rest was not read. When we retested on 2026-10-07, both on Workers AI and through OpenRouter, Clef read and billed the whole text, so the formula now applies without a cap.
² Fitted on 2026-10-10, the day after launch, on the same nine requests.
The formulas are for English prose. Code, JSON and Chinese, Japanese or Korean text produce more tokens per character on every model.
Three workloads, priced
At list price, per 100,000 evaluations:
| Workload | Jev | Clef-flash | Clef | Decider | Luna Decisions | Decision-1 |
|---|---|---|---|---|---|---|
| Router: 1 question, 300 characters | $1.48 | $1.71 | $4.56 | $0.53 | $2.10 | $0.35 |
| Ticket triage: 3 questions, 1,500 characters | $2.66 | $5.53 | $14.74 | $3.94 | $6.84 | $1.47 |
| Document review: 5 questions, 8,000 characters | $7.47 | $17.35 | $46.27 | $27.76 | $20.21 | $6.21 |
The cheapest model changes with the shape of the request. Decider is second in the router case and loses the document case to everything but Clef. Microsoft-Decision-1 is the cheapest in all three, but it was measured a day after launch on an Azure endpoint with tight rate limits, and its accuracy only matched Jev’s, so cost is not the only thing to check.
Sources
- List prices. TypeSafe, Cloudflare Workers AI and OpenRouter (vendor)
- Token formulas. Our calls on 2026-10-04 (Clef, Workers AI; cap retested 2026-10-07) and 2026-10-07 (Decider and Luna Decisions, OpenRouter) and 2026-10-10 (Microsoft-Decision-1, OpenRouter) (ours); Jev from What One Jev Evaluation Costs (ours)
- How these models compare on accuracy. Decision models compared (ours)
Frequently asked questions
- Which decision model is the cheapest?
- In our tests, Microsoft-Decision-1 on all three workloads below: $0.042 per million input tokens like Jev, but fewer tokens per call, and it reads the text once. For one question per call Perplexity Decider ($0.04 per million) is close behind; for several questions on the same text it falls back, because it reads the text again for every question. The calculator shows the numbers for your workload.
- Why do token counts differ between models with similar prices?
- Tokenizers and prompt formats differ, and one model bills differently. We measured: Jev ≈ 260 + 0.163 × characters + 43 × questions; Clef and Clef-flash ≈ 29 + 0.168 × characters + 111 × questions, for the whole text; Luna Decisions ≈ 22 + 0.163 × characters + 139 × questions; Decider ≈ questions × (84 + 0.163 × characters).
- Are output tokens charged?
- No. All five decision models bill input tokens only; the answer, a few probabilities, is free.
- How does JevStation price these models?
- Per evaluation in credits, not per token: 1 credit for a standard evaluation (up to 8,000 characters and five questions) and 3 for a large one, doubled for Clef and Perplexity Decider. Credits come in one-time packs, and the first 50 are free on sign-up.
- Is going direct cheaper?
- At high volume and list price, usually yes: you pay only tokens. You also write and run the integration for each vendor, and Jev direct needs TypeSafe waitlist access. JevStation trades a per-call margin for one key, one request format, a playground, batch runs and refunds on failure.
Related
- Decision modelsJev, Clef, Clef-flash, Perplexity Decider and OpenAI Luna Decisions on the same inputs: accuracy, calibration, cost and latency.
- Jev vs DeciderPerplexity Decider against Jev: same contract, measured accuracy, calibration, latency, and per-question billing.
- Decisions API alternativesOpenAI Decisions API without the waitlist, plus Jev, Clef and Perplexity Decider: measured accuracy, cost, and how to port a request.
Make it part of your pipeline
Sign up for 50 free credits, save your own question sets and call the same evaluation from the API.