Free
$0
100K tokens + 5 free runs every day
- Five successful free runs per UTC day after email verification.
- Tokens never expire
- Workbench, batch + API share one balance
- Failed requests are not charged
PRICING
One balance for the workbench, batch jobs and API. Pay once for tokens that never expire.
No subscription or deadline. Larger packs lower your cost per decision.
$0
100K tokens + 5 free runs every day
$4.90
8M tokens · one-time purchase
Price per 1M input tokens: $0.61
≈ 16,000 requests at 500 input tokens each
Try a first workflow
$19.90
50M tokens · one-time purchase
Price per 1M input tokens: $0.40
≈ 100,000 requests at 500 input tokens each
Regular experiments
35% lower cost per token than Starter
$49
180M tokens · one-time purchase
Price per 1M input tokens: $0.27
≈ 360,000 requests at 500 input tokens each
Production pilots
56% lower cost per token than Starter
$99
450M tokens · one-time purchase
Price per 1M input tokens: $0.22
≈ 900,000 requests at 500 input tokens each
Higher-volume workloads
64% lower cost per token than Starter
Your five free successful runs are used first each UTC day, subject to the site-wide daily capacity shown in the workbench. After those are used, successful Playground and API requests deduct the Jev input tokens reported by the model from your shared balance. Requests that fail before a result is saved are refunded. The 100K free token grant is available in the same balance and never expires.
Request counts are illustrations; actual usage depends on the model-reported input tokens for each request. Free daily runs are separate.
These are JevModel service pack prices, separate from the model provider’s direct rate. View model pricing ↗
Need API access? Create a key in account settings and call POST /api/v1/systemone.
The five free successful runs are used first each UTC day, subject to site capacity. After that, only successful requests consume the model-reported Jev input tokens. Output is free.
No. These are one-time packs. Unused tokens stay in your account without a subscription.
No. Each row is a normal Jev run with the same free-run and token rules as the workbench and API.
The cards estimate requests at 500 input tokens each. Longer input and more questions use more tokens; the actual model usage determines the charge.
Secure checkout is provided by Stripe. Prices are in USD.
JevModel is an independent service and is not affiliated with TypeSafe AI.