Jev pricing
TypeSafe charges $0.042 per million input tokens for Jev, and nothing for output. What that means per decision, and how DecisionNode's prices compare.
Facts checked
Jev's price
Jev is billed per input token at $0.042 per million, which TypeSafe also writes as $42 per billion. Output tokens
are free. Every answer reports usage.input_tokens, so you can see exactly how a request was counted.
- Free tier: none listed in TypeSafe's documentation.
- Rate limits: 100,000 tokens and 80 requests a second per key, with higher limits on custom and enterprise plans.
- Images: not accepted. If your inputs include images, another model or service has to turn them into text first, and that is billed separately.
Source: TypeSafe's models page, checked October 12, 2026.
What a million decisions cost
At list prices, for four request sizes. A short ticket with three questions is around 300 tokens; a long document is several thousand.
| Input tokens a decision | TypeSafe Jev 1.13 | DecisionNode MoE | DecisionNode Flash | MoE, batch | Flash, batch |
|---|---|---|---|---|---|
| 100 | $4.20 | $4.20 | $3.50 | $2.94 | $1.75 |
| 300 | $12.60 | $12.60 | $10.50 | $8.82 | $5.25 |
| 1,000 | $42.00 | $42.00 | $35.00 | $29.40 | $17.50 |
| 4,000 | $168.00 | $168.00 | $140.00 | $117.60 | $70.00 |
Every column assumes the same token count. Each model counts tokens with its own tokenizer, so measure your own requests: both APIs report input tokens on every answer.
DecisionNode's prices
| Item | DecisionNode MoE | DecisionNode Flash |
|---|---|---|
| Text, per 1M input tokens | $0.042 | $0.035 |
| Images and video, per 1M tokens | $0.09 | $0.08 |
| Batch text, per 1M input tokens | $0.0294 (30% off) | $0.0175 (50% off) |
| Fine-tuned model, per 1M input tokens | $0.062 | $0.055 |
| Output tokens | Free | Free |
- Free tier: 50 DecisionNode Flash requests, 20 DecisionNode MoE requests and 10 booster calls a day. On the first workspace you create, renewed at 00:00 UTC every day.
- Boosters (specialist readers for images): age $0.024, ai_generated $0.006 a call.
- Batch jobs return within hours and take up to 10,000 requests a job.
Source: decisionnode.com/docs/pricing, checked October 12, 2026.
Jev pricing questions
Does Jev have a free tier?
TypeSafe's documentation lists no free tier as of October 2026. DecisionNode's free tier is 50 DecisionNode Flash requests, 20 DecisionNode MoE requests and 10 booster calls a day, on the first workspace, renewed daily.
Are Jev's output tokens billed?
No. TypeSafe charges per input token only. Jev generates no text, so the answers themselves are not billed. The same is true of DecisionNode.
Is DecisionNode cheaper than Jev?
DecisionNode MoE costs the same as Jev per input token, $0.042 per million. DecisionNode Flash costs $0.035, about 17% less, and batch jobs cost $0.0294 (MoE) and $0.0175 (Flash). Tokenizers differ, so the cheapest option for you depends on how each model counts your requests.
How do I estimate my Jev bill?
Send a sample of your real requests and read usage.input_tokens from each answer. Multiply the average by your monthly request count and by $0.042 per million.
Run your own decisions on DecisionNode
Send text, JSON or images with typed questions and get probabilities back. Free tier: 50 Flash and 20 MoE requests a day.
Try DecisionNode