ModelPriceWatch.com
Last scan 2026-10-18 Models tracked 292 Providers 40 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

Jev 1.14

by TypeSafe AI

Current new · 4d budget cheap tier
Availability: Sold only by TypeSafe, on a single endpoint (POST /v1/systemone). The versioned id is jev-1.14.0. The jev-preview alias points to it; jev-latest, the default in TypeSafe's SDKs, still resolves to jev-1.13.0 and TypeSafe says it will move to 1.14 about a week after the October 14 release. TypeSafe says it will give 30 days' warning before this version is formally deprecated. No reseller carries it and it is not on OpenRouter. TypeSafe does not print rate limits on the models page: they are set per organization and shown in the developer console, with higher limits on custom and enterprise plans.

Today's price · per 1M tokens

Input

$0.042

per 1M tokens

Output

$0

per 1M tokens

Blended

$0.032

blended $/1M — 3:1 weighted input:output

How this price is scoped: First-party list price. TypeSafe makes Jev and sells it only through its own API, so its models page is the rate card. That page prints one price for Jev 1.14, the same as for Jev 1.13: $42 per billion tokens, which is $0.042 per million — the same input rate in two units, not an input/output pair. Under the table TypeSafe states that you pay per input token and that output tokens are free, so the $0 output here is the maker's own published rate, not a promotion and not a missing number. LiteLLM's public price table lists jev-1.14.0 at the same input rate with free output.

Source: official TypeSafe AI pricing · read MODELPRICEWATCH.COM · 2026-10-18

Price receipt

We read TypeSafe AI's own pricing page on and found Jev 1.14 listed with a price on the same page — that read is where the number above comes from, recorded as a new model. We keep the page text we read; its content hash is 865fe7cdd3.

Overview

TypeSafe AI's newest Jev release, available since October 14, 2026, and like Jev 1.13 a System One model: you send a block of state plus typed questions — yes/no, choice or score — and get structured answers with calibrated probabilities back, instead of generated text. TypeSafe describes it as small improvements across the board with the largest gain in multilingual support, and it is the only Jev that answers with a bounding box; image input and a longer context are available on both versions behind opt-in headers TypeSafe labels experimental. The billing is unchanged: input tokens only, at $0.042 per 1M tokens ($42 per 1B, the unit the rate card leads with), with output tokens free. One request carries a 64,000-token budget covering the state and every question together, with a 32,000-token limit on the state plus the single longest question; the long-context header raises both to 262,000 tokens at the same per-token price.

Capabilities

struck through = not supported
Input 1/4
Text ✓ Image Audio Video
Output
Text ✓
Features 0/9
Prompt caching Reasoning Coding Fast inference Long context Open weights Multimodal Web search Realtime

Specifications

Provider
TypeSafe AI
Context window
64K tokens
Modality
text
Parameters
Proprietary
Open source
No — proprietary
Released
Status
Current
Last updated
Tags
budget

Availability verified: — listed on TypeSafe AI's own page