Jev Pricing & Token Cost Reference
Jev is billed at $0.042 per million input tokens ($0.000000042 per token, or $42 per billion tokens) with $0.00 (free) output tokens.
Because Jev produces mathematical probability distributions rather than autoregressive prose, output tokens are zero-rated across all verified providers. Billing is based entirely on the combined token length of the shared state and the evaluation questions.
Verified Provider Pricing Matrix
Status: Verified September 2026| Provider | Input Price / 1M Tokens | Output Price / 1M Tokens | Minimum Charge | Supported Route |
|---|---|---|---|---|
| TypeSafe Direct API | $0.0420 | $0.0000 (Free) | None (usage-based billing) | POST /v1/systemone |
| OpenRouter | $0.0420 | $0.0000 (Free) | Standard OpenRouter credit balance | POST /api/v1/systemone |
Provider Verification Note: Only providers with officially verified availability and published rate cards are listed. Unverified third-party routers are omitted.
Context Window & Token Budget Architecture
Sources: TypeSafe Docs & OpenRouter Model DirectoryApplies to the cumulative sum of tokens in the shared state block plus all questions included in the request. If you batch 20 questions against a 20k-token state, the total request must fit within 64,000 tokens.
A separate hard ceiling applies to the shared state plus the single longest individual question. Regardless of how few questions are passed, this pair cannot exceed 32,000 tokens.
OpenRouter documents Jev 1.13 as having a flat 32,000-token context window in its model catalog. While TypeSafe exposes the 64k total request budget on direct endpoints, OpenRouter publishes a 32,000-token ceiling for unified compatibility.
Token-Cost Calculator
Estimate your raw token charges by inputting your projected input token volume:
Token Cost Calculator
Evaluates raw token expense based strictly on published provider rates.
Calculation Guardrail: This tool calculates strictly from user-specified token volume and the published rate of $0.042 per 1M input tokens. It does not fabricate representative decision sizes or project unmeasured application costs. Token counts per request depend entirely on your payload state and questions.
Structural Economic Comparison
Autoregressive Output Billing vs. Zero-Output Classification Economics
General-purpose autoregressive LLMs charge separately for input tokens and variable generated output tokens. In generative pricing models, output tokens are typically billed at 3× to 5× higher rates per token than input tokens, and total cost scales with how verbosely the model formats its response.
In contrast, Jev eliminates output billing entirely ($0.00 / output token). Because Jev returns structured evaluations (categorical Choice strings, continuous Score floats, or binary Noul probabilities) rather than autoregressive prose sequences, billing is calculated exclusively from the input token volume at $0.042 per million input tokens ($0.000000042 per token).