All models
TypeSafe AI
Jev 1.13
TypeSafe AI's first System One model for fast, typed choices, scores, and probabilities that software can use directly.
Plain-English overview
What Jev 1.13 actually is
Jev is a specialist decision model, not a conversational language model. An application sends text or structured state plus one or more predefined questions, and Jev returns bounded values with probabilities instead of drafting an answer token by token.
That design makes Jev interesting for high-volume classification, routing, risk scoring, agent checks, and other places where code needs a small decision rather than a paragraph. TypeSafe also exposes uncertainty, which lets a workflow accept confident decisions, escalate ambiguous cases, or fall back to a larger model.
Good fit for
- High-volume classification and request routing
- Confidence-gated agent checks and guardrails
- Narrow semantic judgments inside search and RAG workflows
Category comparison
The facts that matter for decision models
These are provider-published specifications, not Cody benchmark scores. Follow the linked sources for current limits and endpoint-specific exceptions.
- Input price
- $0.042 / 1M input tokens; output freeCurrent provider price for the text or structured state submitted to the model, including whether decision outputs are billed.
- Input capacity
- 64K tokens total; state plus longest question up to 32KMaximum documented state and question capacity for one request.
- Decision types
- Choice, Score, and Noul probability questionsThe bounded question forms the model can answer, such as choices, scores, or probabilities.
- Output contract
- Typed JSON decisions defined before inferenceHow the response is constrained for direct use by application code.
- Uncertainty
- Probability distributions; Choice and Score also include confidenceWhether the endpoint exposes probabilities, confidence, or another calibrated signal.
- Where to use it
- Early-access API and Playground; Python and JavaScript SDKsCurrent API, SDK, playground, and access-stage information.
API and provider access
Where to get Jev 1.13
TypeSafe AIJev quickstartOfficial Python and JavaScript SDK setup, endpoint, and example requests.TypeSafe AIJev model and pricingCurrent model ID, aliases, token price, context limits, and early-access rate limits.TypeSafe AIRead the Jev announcementOfficial architecture, benchmark claims, intended use cases, and launch caveats.
Availability
Regions and access stage
Early access through TypeSafe's Playground and API. The stable alias jev-latest currently resolves to jev-1.13.0.
TypeSafe's privacy policy says its services are hosted in the United States; the model documentation does not publish regional endpoint choices.
Check live availabilityData and training
The route matters.
TypeSafe says it does not train or fine-tune on customer inputs. Its privacy policy does not promise a fixed zero-retention period and says services are hosted in the United States, so sensitive workloads should confirm contractual retention and processing terms.
This is a concise reading of the cited provider material, not legal advice. A third-party gateway can have different storage, routing, training, and residency terms from the model maker's direct API.
Read the provider policyFrequently asked questions
Jev 1.13 FAQ
What is Jev 1.13?
TypeSafe AI's first System One model for fast, typed choices, scores, and probabilities that software can use directly. Jev is a specialist decision model, not a conversational language model. An application sends text or structured state plus one or more predefined questions, and Jev returns bounded values with probabilities instead of drafting an answer token by token.
When was Jev 1.13 released?
Jev 1.13 was released on September 15, 2026 according to the cited provider materials.
Where can I access Jev 1.13?
Early access through TypeSafe's Playground and API. The stable alias jev-latest currently resolves to jev-1.13.0. The access routes listed in this guide are TypeSafe AI.
How much does Jev 1.13 cost?
$0.042 input / 1M tokens · output free. TypeSafe lists Jev 1.13 at $42 per billion input tokens. Decision outputs are not billed. Early-access limits and pricing can change.
Where is Jev 1.13 available?
Early access through TypeSafe's Playground and API. The stable alias jev-latest currently resolves to jev-1.13.0. TypeSafe's privacy policy says its services are hosted in the United States; the model documentation does not publish regional endpoint choices.
Is my Jev 1.13 API data used for training?
TypeSafe says it does not train or fine-tune on customer inputs. Its privacy policy does not promise a fixed zero-retention period and says services are hosted in the United States, so sensitive workloads should confirm contractual retention and processing terms. The policy belongs to the provider route and account terms, so verify it again before production use.