Explore Jev listings

Convai Innovations open decision stack: Apache-2.0 ModernBERT checkpoints, Router mode, and the upstream Python SDK.

Vercel's agent framework with Jev wired into the experimental evaluate path, so scoring stays typed.

Jev picks the browser op and DOM target in one shot; a tiny LLM only shows up when something needs typing.

Semantic ifs on a home 3090: open models doing structured branches without pretending to be TypeSafe.

Contrastive-LM Apache-2.0 8B System One style model built on Qwen3-8B.

Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dro

agent-desktop (github.

Train a small model that chooses among a changing list of text options, one probability per option in a single pass.

Vercel Labs terminal CLI that can point evaluate at Jev instead of another vibes-based judge.

One AI trade call every Monad block on Kuru MON-USDC, with Jev picking the move instead of a market essay.

hermes-jev-skills (github.

Official Claude Code / Codex skill pack for Jev primitives, patterns, and evaluation shape.

The official TypeScript SDK: inferred answer types, System One primitives, and fewer any-shaped regrets.

Every field in POST /v1/systemone, spelled out for engineers who ship gates, not grad students writing RFC fanfic.

Hosted typesafe-ai/jev for AI SDK evaluate calls, no TypeSafe waitlist required.

classifier.

Ploy is a marketing site product that personalizes funnel pages without running a full A/B test program for every headline tweak.

TanStack AI is the TanStack team's AI SDK surface.

Dub (dub.

PixelML (pixelml.

Hypit (hypit.

DocJev (github.

Browserbase builder Shubhankar Srivastava rebuilt a FIFA-style match as a love letter to real-time System One loops.

Jev as a Judge (judge.

Jev Arcade at jevarcade.

Virlo (virlo.

Apache-2.0 clef and clef-flash decision models on Workers AI with open weights on Hugging Face.

Perplexity Apache-2.0 27B decision weights plus hosted Decisions API.

Hosted gpt-6-luna decisions beta with image input and predicate-shaped requests.

Open-source AI Trading OS: research, backtest, paper/live execution, Agent Gateway + MCP; optional TypeSafe Jev pre-trade filter on entries.

Kev is for engineers who want open Jev-style decisions on GPUs they run. Jared Palmer's Apache-2.0 family (github.com/jaredpalmer/kev) serves Choice, Score, and Noul from Qwen checkpoints at 0.8B, 4B, 9B, and 27B, with README eval tables against hosted Jev.

Android chat co-pilot overlay: Jev judges intent and danger on screen text, ranks three reply drafts, fill-only never auto-send.

NanoJev (github.

Nimble (github.

Open multilingual non-autoregressive System 1 engine: a script-aware Router picks English, multilingual, or typed-decisions checkpoints, then one forward pass returns Choice, Score, and Noul.

Apple Silicon Core ML and Neural Engine runtime for open-weight Laya: short typed decisions, PyPI laya-coreml, Hugging Face bundles, Snake demo.

feder-cr/jev CPU binary: OpenVINO int8 jevos-v4 serves POST /v1/systemone on localhost in tens of milliseconds.

jevgrep is for coding agents that need repo search, not another line grep. The jg CLI from dzhng/jevgrep uses Jev to walk folders, files, and declarations, then prints verbatim excerpts. A README eval on 10 Python SWE-bench tasks kept 8 of 10 solves and measured lower total agent cost than search without it.

Mapika open System One family on Qwen3.5: one forward pass, Choice/Score/Noul, POST /v1/systemone via decider-ai on PyPI, weights on Hugging Face.

Apache-2.0 non-autoregressive System One model: one forward pass, order-invariant option scores, sub-25 ms local claims.

Computer use for about $0.

GitHub Next Bun server: local POST /v1/systemone over oMLX DiffusionGemma chat JSON, TypeSafe SDK compatible, README honest on calibration.

Hippo Memory (hippo-memory.

Rust coding agent harness: Jev routes reasoning consults, effort, and utility summaries; fail-open; Distill owns tool permissions.

LLM plus Jev alloy: JevSift, JevLoop, and JevLong over a CLI JSON protocol so the host agent keeps the session.

macOS menu-bar companion: default JEV mode picks local controls for clicks; Gemini mode for voice and writing. Keys in Keychain.

OpenBMB MiniCPM-OCR Jev: 2B Apache-2.0 scores OCR regions 0 to 1 for human review gates and best-of-N parser selection.

SGLang JSON Schema generation on existing AR LLMs: enums, numbers, depends_on graphs, JevBench 228/231 with thinking. Not a System One server.