Jev 1.14.0 (TypeSafe AI)

Composite (secondary)

70.2 · Not ranked on this board

Cost and measurement conditions

$0.032 per 1,000 decisions (estimated)

TypeSafe AI public list price, USD 0.042/M input, output free: the "Current models" table of TypeSafe's models page (docs.typesafe.ai/models, fetched 16 Oct 2026 22:37:44Z, saved copy sha256 936da179...) lists jev-1.14.0 at "$42 / $0.042" per Btok / per Mtok, the same price as jev-1.13.0, and says "You pay per input token. Output tokens are free."; Jev 1.14 is "available from October 14, 2026". Counted as the manufacturer's standard launch list price under interpretation I-1 (the reading that admitted kushal-gemma4-31b-it-autoloops, SubQ Snap, d1-ultra, Sage 1.4, Nimble, kettei and Celeris-1 Decision), through a scoped re-freeze of the frozen pricing module that adds this one key and nothing else, with a zero-difference proof for every other registry entry; jev-1.13.0 is priced through the same launch-list set at the same tariff. On the same page the alias jev-latest ("the most recent stable, official release") still points to jev-1.13.0, and jev-preview ("the most recent release, official or not") points to jev-1.14.0; the versioned id is listed with its price, accepted by the API and was the served model on every row. ESTIMATE in the scorer's sense: the tariff x this row's OWN provider-reported tokens (1,134,053 input tokens over the 1,485 answered items; the 71,566 reported output tokens are free under the tariff), divided by the 1,485 items of the common cost basis the scorer uses; the 15 requests refused with HTTP 400 report no usage and count as zero tokens and as wrong answers (TypeSafe's docs do not say whether refused requests are billed). USD 0.032074 per 1,000 decisions on the common basis (USD 0.031753 over all 1,500 items), 0.497x the official cost cap. Not a bill and not a quote. COMMON COST BASIS (v1.6.30): cost per 1,000 decisions is computed on the common 1,485 items of this draw, leaving out the 15 very long items (at least 76,358 input tokens as counted by the rows that answered them). They are the 15 items the Jev reference refused with max_tokens_exceeded, which the other rows that did not answer them refused or failed on; their tokens count for no row, whether the row refused or answered them, and a refused item still counts as a wrong answer for Intelligence. This row: USD 0.032074 per 1,000 decisions (0.497x the official cost cap USD 0.064595), against USD 0.031753 over all 1,500 items (v1.6.29).

p50 latency: 0.2 s

none (API)

Vendor-hosted TypeSafe AI API (api.typesafe.ai), model `jev-1.14.0` (served model id `jev-1.14.0` on all 1,500 responses); latency measured from Sandy (Germany) on the official per-item latency subset without API adjustment. Full S1200+P300 draw of `v1.6-regular-20261012` through the unchanged `run_api_full_v161.py` and the unchanged `jevbench.adapters.typesafe`; the full-set exposure gate logged all 1,200 sealed items to provider `typesafe` before the first request (custody #20474). Native probabilities on all three question types. 15 of 1,500 requests returned HTTP 400 max_tokens_exceeded (the 15 long items of this draw; per TypeSafe's API reference the request is longer than the model's context length, 64k per request by default); the runner never re-sends a failed item, so they are scored as wrong in the full 1,500 denominator. TypeSafe documents an opt-in long-context mode (request header TypeSafe-Danger-Zone-Long-Context: true, up to 262k tokens at the same per-token price); this run used the default 64k context, as every hosted row of this draw used its default, so the 15 longest items were refused. 12 answered rows return probabilities that sum to 0.99 (the API rounds to two decimals); the unchanged scorer renormalises distributions within its 2e-2 tolerance, as it does for the published jev-1.13.0 raw, which has the same two-decimal property. Hosting implementation and deployed weights were not independently inspected.