Context
64K
Max output
—
Input / 1M
$0.042
Output / 1M
Free
Cached / 1M
Free
Cutoff
—
About this model
TypeSafe Jev 1.13 System One model. Evaluates typed questions (choice, score, noul) against a state and returns calibrated answers with probabilities and confidence. Not a text generator: send the input as the user message and the questions as configuration.questions.
Best suited for
- Classification and intent routing
- Scoring against a rubric
- Yes/no checks with a probability
- Guardrails, validation and re-ranking
Supported parameters
questionsMap of typed questions sent per request as configuration.questions. Each entry: { type: 'choice' | 'score' | 'noul', instructions, criteria? }.
Use this model
OpenAI-SDK compatible, with gateway fallback and routing across providers.
Other Typesafe models
| Model | Context | Max output | Input | Output | Capabilities |
|---|---|---|---|---|---|
| jev-1.14.0 Chat | 64K | — | $0.042 | Free | — |
| jev-latest Chat | 64K | — | $0.042 | Free | — |
Pricing & provider details
Typesafe
typesafe/jev-1.13.0
- Input · per 1M
- $0.042
- Output · per 1M
- Free
- Cached input · per 1M
- Free
- Context window
- 64,000 tokens
- Max output
- —
- Knowledge cutoff
- —
- Auto-router
- Not supported
Start building today
Route jev-1.13.0 — and every other model in the catalogue — through one endpoint, with failover built in.