Article
CommunityJev Guides & Articles
Jev 1.14 Rollout Notes
Jev News's rollout story for Jev 1.14: OpenRouter listed typesafe/jev-1.14 while TypeSafe's jev-latest alias still returned jev-1.13.0 at 06:16 UTC, with the desk's own API calls recorded across the day.
- Format
- Article
- Reading time
- 3 min
- Published by
- Community
- Author
- Jev News
- Last verified
- Oct 2026
Highlights
- At 06:16 UTC on October 15, every jev-preview call named jev-1.14.0 and every jev-latest call named jev-1.13.0.
- Three Chinese payouts tickets scored frustration 1.27, 1.32 and 1.32 on jev-1.14.0, mean 1.30 at 0.54, against 1.54 at 0.31 on jev-1.13.0.
- Language came back at 1.0 on English, Chinese, Spanish, Japanese, French, German and Korean on both models.
- OpenRouter listed typesafe/jev-1.14 at $0.042 per million input tokens, output free, a 262K context and a TypeSafe P50 of 0.15 s.
- The alias table still named jev-latest as jev-1.13.0 and jev-preview as jev-1.14.0, with a deprecation policy promised first.
Watch out
A community-desk story that mixes TypeSafe's blog and models page with the desk's own small API runs: the frustration comparisons are three-request samples, the desk did not fill a 262k context or call OpenRouter, and the site says its figures can go stale. It shows no licence or open weights for the models.
Questions about this entry
Short answers drawn from the linked project or page.
At 06:16 UTC on October 15, every jev-latest response still named jev-1.13.0 while every jev-preview response named jev-1.14.0, even though OpenRouter had already listed typesafe/jev-1.14 earlier that UTC day.
The listing showed $0.042 per million input tokens, output free, a 262K context and a TypeSafe P50 of 0.15 s. The desk notes that the latest alias page already pointed at Jev 1.14 and that OpenRouter's models endpoint did not return the new id.
Three Chinese payouts tickets scored 1.27, 1.32 and 1.32 on jev-1.14.0, mean 1.30 at 0.54 confidence, against 1.55, 1.52 and 1.55 on jev-1.13.0, mean 1.54 at 0.31. All are the desk's own calls.
Four payouts sentences in English, Chinese, Spanish and Japanese all came back at language confidence 1.0, and billing stayed the top department on all four. In the larger run, language scored 1.0 on seven languages on both models.
The post, opened October 15, dated October 14 and titled Jev: For the builders AND dreamers, says the gains are small and across the board, with the largest expected in multilingual support, and that releases should generally be strict improvements.
The blog says jev-latest will point at jev-1.14.0 in a week and jev-1.13.0 stays available and is not yet deprecated. TypeSafe intends 60 days of warning before jev-1.13.0 is formally deprecated and 30 days for jev-1.14.0, with a deprecation policy landing first.
Posted October 14 at 21:52 UTC, it says TypeSafe's first public Jev release favored extremely stable infrastructure, while this one is for early adopters, tinkerers and dreamers, with a gate so a caller can opt into a danger zone.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Both Jev 1.13 and 1.14 score 29 of 30 on 30 Turkish buyer and seller messages
X: Microsoft-Decision-1 against Jev on a private eval
X: Vals AI's independent evaluation of Jev
X: An eval suite for System One models
More Jev resources like this
Jev News's desk for TypeSafe's October 14, 2026 Jev Day: Jev 1.14.0 shipped as jev-preview with opt-in images, a Box type and 262k context, plus the desk's dated API checks and what they returned.
Jev Guides & ArticlesDocs#community#news#danger-zone
Community
Chinese-language Jev cookbook with runnable Jupyter notebooks covering the three primitives, 18 official recipes, evaluation method, voice smart-home demo, and local Laya fine-tuning.
Jev Guides & Articles#community#jevbest#catalog-import
Blackwell Systems' adversarial technical due diligence on Jev, built only from public material and live API calls. The author reproduces TypeSafe's published addressee numbers within a point and finds a real but noise-type-specific robustness moat.
Jev Guides & ArticlesArticle#community#press#competition
Community
A free animated course on System One decision models, covering TypeSafe Jev alongside OpenAI Decisions, Cloudflare Clef and AWS Strands Decider: 22 short lessons backed by 199 sources and 14,318 API calls the authors ran themselves.
Jev Guides & Articles#community#github#course
Community
Haple/system-oneJavaScript
Haple/system-oneJavaScriptA Korean Slidev course — 59 slides, 110 minutes plus deep-dive Q&A — with three standalone Jupyter notebooks that take Jev decision models from NLL/Brier/ECE calibration to training a Qwen-based Jev-compatible model with Unsloth and testing its HTTP API.
Jev Guides & Articles#community#github#korean
Community
heury/jevJupyter Notebook
heury/jevJupyter NotebookDeepInfra's survey of the decision-model category: pointer-head architecture, RLCD calibration, latency and cost figures for Jev and its rivals, and a landscape table spanning TypeSafe, AWS, Convai, Kev, OpenAI, and Ollaya.
Jev Guides & ArticlesArticle#community#analysis#architecture
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.