vllm-project/vllm
A vLLM pull request adding a Jev-like structured-decision path on Gemma, cited by testers comparing open alternatives on the Pareto frontier with reflex and DiffusionGemma.
vLLM PR #57250: a Gemma-based Jev-like path →REPOSITORY ACTIVITY
Latest pushes
Latest repository pushes from our saved GitHub snapshot. Each repository has its own metadata check time; the list changes when that snapshot is refreshed.
283repositories with a recorded push
286indexed repository addresses
2 redirected addresses are merged into their current repositories. 1 repository has no usable push metadata in this snapshot. Browse all repositories → · Activity JSON
283 repositories, ordered by last push.
-
-
trycua/cua
A Cua Driver integration that uses Jev to choose bounded computer actions on macOS, Windows and Linux.
jev-use for computer use → -
BerriAI/litellm
The LiteLLM proxy scores older tool results with Jev and replaces low-relevance ones with short removal notices before the request reaches the LLM; a TypeSafe pass-through endpoint is documented too.
LiteLLM: Jev-driven context compaction and pass-through → -
run-llama/liteparse
An OSS library from the LlamaIndex founder: give a document plus natural-language category rules and Jev predicts the category (classify) or the boundaries between sub-documents (split). OCR backends: liteparse (fast, free) or LlamaParse for complex layouts.
DocJev: document classification and splitting with Jev → -
AppitStudio/awesome-jev
A community directory of Jev projects kept as one markdown file per project (TipTour was added within hours of its release).
AppitStudio awesome-jev: community project directory → -
dzhng/jevgrep
A CLI that uses Jev to judge relevance across repository folders, files and declarations, returning file leads and verbatim source excerpts to a coding agent.
jevgrep: semantic code search for coding agents → -
pydantic/pydantic-ai
Pydantic AI merged a TypeSafe model provider: `TypeSafeModel('jev-latest')` with confidence thresholds, a FileUnderstanding capability and fallback to an LLM for tool calls; docs cover jev-latest, jev-preview and pinned releases.
Pydantic AI: first-class TypeSafeModel → -
braintrustdata/braintrust-sdk-python
Braintrust added Jev as a scorer with a confidence-threshold pattern (accept above 0.95, route 0.70–0.95 to an LLM judge, flag below 0.70 for humans) and auto-instrumentation of TypeSafe client calls in its Python SDK.
Braintrust: Jev as a typed scorer for agent responses → -
wfzyx/von
`pip install von`: a non-autoregressive local decision model claiming wire-protocol compatibility with Jev (`von serve`), with ModernBERT or Laya-421M backends; Apache-2.0.
von: a CPU-only local decision model → -
jkudish/jev-browser
A headless-browser agent uses Jev to select actions from real page elements and exposes the same navigation loop through MCP, CLI and a library.
Jev Browser: Playwright agent through MCP, CLI or library → -
jkudish/jev-mcp
A proof-of-concept MCP server exposing three tools: verify a claim against evidence, screen text for injection or relevance, and rank candidates semantically.
jev-mcp → -
leepokai/jev-guard
Risk-scores every tool call (rm, curl, bash…) with session context into deny / ask / allow, flags prompt injection in tool results and checks skills and plugins; works with Claude Code, Codex and others.
jev-guard: auto mode for every coding agent → -
Arize-ai/phoenix
A Python package that traces TypeSafe System One calls as OpenInference LLM spans (request, response, model, token counts) for Phoenix and compatible backends; a JS package is documented in Phoenix.
OpenInference instrumentation for TypeSafe (Arize) → -
ollaya-dev/ollaya
A local daemon and CLI for downloading and running open decision models behind a TypeSafe-compatible API. It includes an MCP interface and model definitions for typed Choice, Score and Noul workflows.
Ollaya: pull and serve open decision models locally → -
superagents-lab/jev-search
Jev chooses sources and query settings, then re-ranks results returned by search engines. It does not generate an answer.
Jev Search → -
mizorewww/laya-coreml
Native MLX and Core ML runtimes for Laya typed decisions: `pip install 'laya-mlx[demo]'`, no text generation, no PyTorch, no cloud API. A Rubik's-cube demo runs 100 moves locally.
laya-mlx / laya-coreml: Laya on Apple Silicon → -
mizorewww/laya-mlx
Native MLX and Core ML runtimes for Laya typed decisions: `pip install 'laya-mlx[demo]'`, no text generation, no PyTorch, no cloud API. A Rubik's-cube demo runs 100 moves locally.
laya-mlx / laya-coreml: Laya on Apple Silicon → -
superdesigndev/treg
Treg ('OpenRouter for agent tools') with Jev on the decision step across fraud detection and signup screening, buying-signal triage and viral-content monitoring.
Treg + Jev: three workflows, $8K/month saved → -
nokia-applied-research/AnyJev
A Python library that turns open causal models into Choice, Noul and Score decisions with access to logits or hidden states.
AnyJev: typed decisions from open models → -
AnotiaWang/awesome-decision-models
Another curated list of Jev / System One applications, libraries and resources, linked from the 'Jev 101' article.
AnotiaWang/awesome-jev → -
langfuse/langfuse-docs
A tutorial that replaces an LLM-as-judge with Jev to score Langfuse traces, with a full Python example detecting user disagreement per conversation turn.
Langfuse: using Jev for evals → -
yibie/awesome-jev
56 entries in twelve use-case categories plus links to the Hacker News and Reddit launch discussions.
yibie/awesome-jev → -
jaredpalmer/kev
A family of tiny Jev-like decision models — 0.5B at launch, now 0.6B, 4B and 8B on Qwen3.5 — using the same LoRA-plus-pointer-head recipe with block-causal masking so questions run in parallel; the HTTP API mirrors System One, so the official SDK works with only a base_url change.
kev: a tiny Jev-like you can train on a MacBook → -
itsmostafa/system-one-connector
An MCP connector that gives any MCP-capable agent direct Choice / Score / Noul calls to Jev; the Chinese roundups call it the first thing to install after getting an API key.
typesafe-mcp: Jev in Claude Code, Claude Desktop, Codex and Pi → -
TypeSafeAI/jev-harness
A research-stage coding harness where an LLM proposes an action, Jev answers four bounded questions and code records evidence for the host to review.
Community jev-harness: proposal review with four questions → -
Arize-ai/openinference
A Python package that traces TypeSafe System One calls as OpenInference LLM spans (request, response, model, token counts) for Phoenix and compatible backends; a JS package is documented in Phoenix.
OpenInference instrumentation for TypeSafe (Arize) → -
0x7067/jev-browse
A browser agent gives Jev a numbered page table and executes its chosen operation and row, using a text model when text entry is required.
jev-browse: Jev-driven Chrome automation → -
reticlehq/reticle
Agents can write code but struggle to understand what they built at runtime; Reticle gives web and desktop apps a Jev-style runtime perception layer so agents know which interface they are looking at.
Reticle: machine-native runtime perception for web and desktop apps → -
kerpopule/hermes-jev-skills
Jev-powered model routing, memory, context compaction, skill selection and computer/browser use for Hermes agents (also Claude Code and Codex), with a routing dashboard and shadow mode that shows what Jev routing would have saved before you switch it on.
hermes-jev-skills: routing, memory, compaction, GUI clicks for Hermes → -
justhalfbit/dsh-plugin-jev-effort-selector
A DeepSeek Harness plugin asks Jev to choose a supported reasoning-effort level for a task and preserves the previous level when the choice fails.
DSH Jev Effort Selector → -
NandhaKishorM/laya
An mmBERT-based 0.3B typed decision model published a year before Jev and rediscovered during launch week; weights and a demo Space on Hugging Face, plus a multilingual variant. Ports to MLX and Core ML followed within days.
Laya: the open decision model that predates Jev → -
RiskAverseTech/toolgate
Screens agent tool calls for destructive, exfiltration, privilege-escalation and off-task actions using Jev probabilities, ships as an npm package and a Claude Code PreToolUse hook, and logs an audit trail.
toolgate: a tool-call firewall for agents → -
TypeLLM/TypeLLM
An open-sourced decoding engine for autoregressive LLMs inspired by Jev: one output token per decision, no out-of-schema outputs, KV-cache reuse; demoed on SGLang with Qwen 27B.
TypeAR: Jev-inspired type-safe decoding → -
ekzhang/openjev-sglang
A server that implements the TypeSafe/Jev HTTP API on Qwen3.6-35B-A3B with SGLang: prefill-only, one extra token per question, radix caching, deployable on Modal.
openjev-sglang → -
featherless-ai/simple-jev
SimpleJev.ai from Featherless: a fully open-source library that takes any Hugging Face model and 'Jev-ifies' it behind an API endpoint, with public no-key endpoints, vision via chat-format images (Gemma, Qwen) and an RFDT tuning recipe to distil your prompts into a smaller production model.
simple-jev → -
spring-ai-community/spring-ai-typesafe
A Java System One client plus Spring AI modules for Jev judging, guardrails, retrieval filtering and reranking, and tool selection.
Spring AI TypeSafe: Java SDK and Jev integrations → -
kishida/llama.cpp
A llama.cpp fork with a `jev` branch and docs/jev.md that runs Jev-style typed decisions on local models; the author's size-by-size picks are Qwen3.8-27B, Gemma 4 12B, Qwen3.5-4B and two tuned 'jwenv' Qwen3 PoCs.
llama.cpp jev branch: local decisions on Qwen and Gemma → -
zhengxuyu/litjev
Turns any off-the-shelf LLM into a Jev-like decision layer and serves the same /v1/systemone API locally — no API key needed.
litjev: /v1/systemone on top of a local Qwen via vLLM → -
kushals256/jevcache
An OpenAI-compatible local cache proxy (`npx @kushalicious/jevcache`): agents that re-ask the same thing get the cached answer when Jev judges the intent identical.
jevcache: skip the LLM call when Jev says same intent → -
cobanov/awesome-jev
A source-backed list of 70+ Jev projects in eight categories, including open reproductions and evaluations; reviewed 2026-09-18 per its README.
cobanov/awesome-jev → -
chy4pro/jev-for-chrome
A Manifest V3 port of jev-ultrafast that lets Jev choose clicks, keystrokes and dropdown actions in the user's existing Chrome tab; a separate text model handles typed content.
Jev for Chrome: in-tab browser actions → -
jammaru/jev-lab
A TypeScript workspace containing Hundred, a town of 100 NPCs, and Jev Shogi, a human-versus-model board game. Each Jev request carries compact state and engine-legal next actions.
Jev Lab: a 100-NPC town and a shogi opponent → -
y0usaf/pi-jev
Jev as a decision layer for the Pi coding agent: a measured tool-call gate plus a jev_ask tool for typed, calibrated answers; distinct from madeye/pi-jev.
pi-jev (y0usaf): a measured tool-call gate for Pi → -
lahfir/agent-desktop
A Rust runtime that lets any agent operate desktop apps through their real UI structure (OS accessibility trees) with stable element refs, from the author of the hook-based routing demo.
Agent Desktop: reliable computer use through OS accessibility trees → -
ainame/swift-typesafe
A port of typesafe-sdk-python to Swift.
swift-typesafe: an unofficial Swift SDK → -
jev-chat/jev-chat-jarvis-mac
A macOS overlay that reads WeChat on screen, runs an optional TypeSafe Jev or local decision model, and prepares reply suggestions for manual sending.
jev-chat-jarvis-mac: WeChat reply assistant → -
iamaamir/system-one
A TypeScript System One runtime provides shared Choice, Noul and Score builders for Jev, Reflex and compatible endpoints, with a companion Pi extension.
System One for TypeScript: provider-neutral decisions → -
AbdelStark/awesome-typesafe-jev
A curated list of official and community TypeSafe / System One / Jev resources, 50+ entries in six categories, with a GitHub Pages site; reviewed 2026-09-17 per its page.
awesome-typesafe → -
AboveColin/HA-Jev
Home Assistant integration: ask the house a typed question and get a number back — Jev answers exposed as sensors, four actions for automations, and a conversation agent for Assist.
HA-Jev: Home Assistant integration → -
coldteadotai/abide
Uses Jev to enforce the project rules linters cannot catch — the judgment calls in a style guide — on a coding agent's output.
abide: make your coding agent follow project rules → -
AkashPriyadarshii/jev-curate
A high-throughput sifter for synthetic and pretraining datasets: stream Parquet and JSONL, filter and score every row with Jev.
jev-curate: sift pretraining datasets at 1,500+ rows/sec → -
rokbenko/quackd
A multi-robot harness CLI (Microduck, Open Duck Mini, LeRobot, XLeRobot, AlohaMini and more) where each robot has an LLM brain and Jev handles the cheaper decision steps; a LeRobot SO-101 setup takes about five minutes.
quackd: one CLI for all your robots, Jev for the cheap steps → -
tamaratran/jev-pruner
A Claude Code and Codex CLI plugin that intercepts Bash output above roughly 10K estimated tokens, chunks it and asks Jev which lines are still needed, dropping progress logs and boilerplate.
jev-pruner: trim tool output before the LLM sees it → -
prasanthj/duckdb-jev
A native DuckDB extension for Jev classification, scoring and semantic predicates from SQL: vectorized input, batching, bounded concurrency, dedup and cross-query TTL/LRU caching. Distinct from colliber/duckdb-jev.
duckdb-jev (prasanthj): native, batched, cached → -
sysevol-ai/CodeNib
CodeNib uses Jev to rerank code-search results and publishes integration docs plus an evaluation on 100 CodeNib Base issues.
CodeNib: Jev reranking for code search → -
GhalebDweikat/winnow
Every tool result is judged by a System One model before it enters Claude Code's context, so noise never gets a chance to bloat the window.
winnow: a calibrated context sieve for Claude Code → -
browser-use/jev-ultrafast
A browser agent where Jev picks the operation and target element from a numbered element table each step, and a small LLM is called only when text has to be typed.
Jev Ultrafast (Browser Use) → -
tin-xai/multimodal-jev-grounding
Answers image questions with option probabilities and bounding boxes; trained on CUB-200-2011 with a Qwen2.5-VL backbone.
Multimodal Jev prototype: option probabilities plus bounding boxes → -
daseinlabs/open-jev
One-pass option scoring with a local Gemma 3 4B via MLX: pre-written options are batched against a cached context; implements the System One API contract and includes a Doom terminal demo.
open-jev (MLX, Apple silicon) → -
awlevin/typesafe-computer-use
A computer-use agent that OCRs the screen, asks Jev to pick the next action, and clicks; the author says the loop generalizes across operating systems.
Computer use on Jev: OCR, classify, click → -
MaxBittker/runebench
Adapts the RuneBench RuneScape agent benchmark to Jev with about fifty bounded actions, a tick-mode controller and a live dashboard.
jevscape: RuneScape benchmark harness for Jev → -
Asymptote-Labs/agent-beacon
Beacon captures sessions across coding-agent harnesses and turns reviewed workflows, corrections and debugging patterns into reusable knowledge.
Beacon: cross-harness memory for coding agents → -
kotoba-lang/typed-decisions
An open typed-decision model on DeBERTa-v3-large answering Choice (up to 255 options), Score and yes/no questions from one forward pass, trained on 18K states and 42K questions.
open-jev-deberta-v3-large (Kotoba Labs) → -
jacks3tr/Jev-Desktop
Desktop computer-use agent released as open source and slated to become part of the author's internal verification workflow.
Jev Desktop: open-source computer use, ~10x faster than native → -
jev-chat/jev-chat-jarvis
An Android chat assistant that uses Jev or a compatible decision endpoint to judge message intent and risk, then offers drafts from a separate generative model.
jev-chat-jarvis: Android reply assistant → -
kieranklaassen/ruby_llm-typesafe
Adds a `:typesafe` provider to RubyLLM 2 with Noul, Choice and Score primitives, schema building and validation, and an offline models catalog.
ruby_llm-typesafe → -
kyotofin/tax-doc-classifier
A Chinese X Article sorting the early ecosystem into things that actually ship: jev-ultrafast, mobile-jev, typesafe-computer-use, AskJev, a Twitter timeline analyzer, a tax-document classifier, jevdetector and newsjack, plus the OpenRouter and DataCamp explainers.
Got Jev — now what? A real deployment list → -
razorback16/openjev
An open, Jev-compatible System One decision server running DiffusionGemma — one of several projects named openjev.
openjev: a Jev-compatible decision server on DiffusionGemma → -
aadilghani1/qc-use
Turns plain-language critical-path tests into bounded UI actions chosen by Jev. The runner checks a fresh screenshot after each action and records a step-by-step result.
qc-use: Jev-assisted UI tests from plain-language steps → -
kavehmz/typesafe-playground
A self-driving simulation built to test Jev: structured sensor state in, typed decisions out (steer, brake, overtake, pedestrians, speed limits); the prompt is published.
Self-driving simulation with typed decisions → -
shhivv/third-hand
A computer-use assistant driven entirely by decision models: sub-second latency, near-zero cost, and it decides what to type without calling an LLM.
third-hand: computer use with no LLM at all → -
Ying-Kai-Liao/jev-browser
Browser automation split in two: an LLM plans the task, Jev makes the concrete in-page choices; shipped as a library, a CLI and an MCP server.
jev-browser: the LLM plans, Jev decides — library, CLI and MCP → -
thruwire/foreman
Foreman watches coding agents and uses Jev to continuously assess progress, completeness, tests, drift and verification, intervening when needed.
Foreman: a supervisor for coding agents → -
ruban-24/switchboard
Model-agnostic router that takes prompt cache into account when Jev picks the model per turn; other System One models planned.
switchboard: a cache-aware decision router for Claude Code and Codex → -
mohamadkhoshnava/ZeroNSFWBot
A Telegram moderation bot that uses Jev for optional text, bio, advertising and language judgments. Image moderation runs through a separate local detector.
ZeroNSFWBot: optional Jev text moderation for Telegram → -
fornhere/hafiza-os
A local Markdown-vault second brain for Linux, macOS and Windows with Claude Code, Codex and Antigravity adapters and optional Mem0 / Jev; one-command setup that also installs Obsidian.
hafiza-os: a sourced second brain with optional Jev → -
Bodila51/grok-bot-jev
A seven-step setup that puts Jev between Grok Bot and its actions: a jev-usage-router skill is consulted before browser, research, retry or extra-bot steps; it starts in shadow mode with logs, then goes active, with a kill switch. Repository included.
Jev as a decision layer for Grok Bot → -
iamvatsalpatel/tiershift
Jev scores each request on eleven dimensions (difficulty, stakes, safety and more) and YAML thresholds map the scores to model tiers; TypeScript and Python SDKs plus an HTTP proxy.
tiershift: policy-bounded LLM cost router → -
jev-chat/jev-chat-windows
A Windows companion that collects visible WeChat messages with local screenshots and OCR, then uses Jev to rank three draft replies for the user to send manually.
Jev Chat for Windows: ranked WeChat reply drafts → -
irfndi/prism-liquidity-agent
An autonomous liquidity-provider agent with auto-rebalancing and backtested strategies (Meteora DLMM first), with Jev in the decision path.
prism-liquidity-agent: autonomous LP rebalancing on Meteora → -
yijunyu/jev-rs
A Rust engine derives typed System One judgments from model log probabilities and exposes them through CLI, MCP and a Jev-compatible HTTP server.
jev-rs: Rust System One scoring and evaluation engine → -
Ryoseiimai/shibuya-machimachi
A Shibuya meetup app that uses Jev to judge whether participants met from distance, floor and elapsed time.
Shibuya Machimachi: a Jev-assisted meetup app → -
Heman10x-NGU/Verdict-open-jev
A non-autoregressive decision engine on ModernBERT (151M) with calibrated uncertainty, an audit of the Jev benchmark, and an in-browser WebGPU playground; Apache-2.0 with PyTorch and ONNX weights on Hugging Face.
openJev-verdict 2.0: a 151M calibrated decision engine → -
oso95/x-scanner
A Chrome extension that labels posts in the X timeline with Jev across five dimensions while scrolling.
x-scanner: label X posts on five dimensions → -
EliaAlberti/jev-rules
A Claude Code plugin that asks Jev which standing rules and codebase maps are relevant to the current prompt or file edit, injects only those, and falls back to all rules if the API is down.
jev-rules for Claude Code → -
nassim-arifette/jevgrep
A TypeScript CLI and local MCP server ask Jev to rank source fragments and return exact excerpts with paths and line numbers to coding agents.
JevGrep: semantic source retrieval through CLI and MCP → -
joaoh82/coffee-under-fire
A React and Three.js arena shooter built around coffee deliveries and Jev-selected NPC tactics. Deterministic TypeScript owns the simulation, while recorded decisions can be replayed without model calls.
Coffee Under Fire: Jev-controlled NPCs in a browser shooter → -
kshetrajna12/reflex
State plus typed questions in, calibrated probabilities out, built on Qwen3.5-4B, with a 650 MB WebGPU browser build on Qwen3.5-0.8B.
reflex: a small open decision model with a WebGPU build → -
altryne/jevify
An independent agent skill for spotting Jev opportunities, designing typed questions, preserving source evidence and testing a small decision workflow.
Jevify: an agent skill for typed-decision workflows → -
typesafe-ai/typesafe-sdk-python
The official Python client (`uv add typesafe-sdk`) with sync and async variants; the quickstart classifies support tickets into billing, technical or other.
typesafe-sdk (Python) → -
walidboulanouar/awesome-jev-use-cases
A public repo collecting things built with Jev plus patterns, documented limits and a list of ideas nobody has built yet.
awesome-jev-use-cases: demos, patterns, limits, unbuilt ideas → -
jerryjliu/docjev
An OSS library from the LlamaIndex founder: give a document plus natural-language category rules and Jev predicts the category (classify) or the boundaries between sub-documents (split). OCR backends: liteparse (fast, free) or LlamaParse for complex layouts.
DocJev: document classification and splitting with Jev → -
TypeSafeAI/clarity-judge
Seven independent writing checks (hedging, filler, passive voice and more), each returning its own verdict, confidence and the driving sentence; runs simulated without a key.
Clarity Judge: seven writing checks → -
mizchi/jev-playground
A MoonBit Jev client and CLI with game, shell-risk and browser experiments and the author's measurement reports.
mizchi/jev-playground: MoonBit client and experiments → -
Tom-R-Main/Footwork
A browser and macOS computer-use harness that checks completion claims against observed state and gates irreversible actions with Jev. Its dual mode also delegates mechanical steps to Jev while a driver handles broader decisions.
Footwork: Jev guards and mechanical computer-use steps → -
bohutang/sift
A Chrome extension that labels every post on X (Substance, Humor, Chit-chat, Promo, Junk, AI-written) plus an AI-probability score, and folds the ones you don't want.
Sift: label every X post with Jev and hide the junk → -
WiktorB2004/llama-index-jev
Two PyPI packages: a `JevRerank` postprocessor using Score and single/multi selectors using Choice for LlamaIndex routing.
LlamaIndex reranker and selectors for Jev → -
bnsd55/jevmlx
Jev-style parallel constrained decisions for any MLX model on Apple Silicon: typed, schema-valid JSON in one forward pass.
jevmlx: parallel constrained decisions for any MLX model → -
aaronshaf/opencode-jev-orchestrator
Keeps OpenCode on one cheap model for a warm cache and lets Jev decide which turns are hard enough to spawn stronger subagents.
opencode-jev-orchestrator: cheap sticky model, Jev escalates hard turns → -
ikermoel/open-alternative-jev
A System One style layer that returns typed, calibrated decisions from any open-weights LLM in a single forward pass (Hugging Face and vLLM), with benchmarks against Jev in the README.
open-alternative-jev: typed decisions from any open-weights LLM in one pass → -
Contrastive-LM/CLM
An independent decision model with published code and model heads, served through a System One-compatible API.
CLM: an independent decision model → -
monteduro/killmyidea
Submit a startup idea and get kill / fix / ship with probabilities.
Kill My Idea → -
RomanSlack/jev-drone
A layered controller: classical vision at 15 Hz, a 50 Hz reflex safety layer, and Jev at about 2.5 Hz choosing tactical maneuvers from JSON scene descriptions.
jev-drone: quadrotor obstacle course in MuJoCo → -
hellogumbo/awesome-jev
A large, actively maintained community directory of SDKs, integrations, agent tools, games, demos and research.
Awesome Jev → -
giuliosmall/pg_typesafe
A pre-alpha PostgreSQL C extension exposing typesafe_noul(), typesafe_classify(), typesafe_score() and batched variants that call the API from SQL.
pg_typesafe: Jev from SQL (C extension) → -
OmniJev/PlayJev
An independent Jev-like model fine-tuned from Qwen3.5-0.8B-Base to choose legal actions from raw game frames. The project publishes model weights, training records and a gallery of ten browser games.
PlayJev: a 0.8B visual decision model for ten games → -
Zaious/jev-capability-atlas
A bilingual (Traditional Chinese / English) evidence map of where Jev holds up and where it breaks, with API-call receipts under suites/. Its thesis: success tracks whether the answer is derivable from the supplied state rather than from outside knowledge.
Jev capability atlas → -
caliber-ai-org/ai-setup
Caliber's agent-configuration CLI includes an optional Jev plugin that scores tool-call history for retention during transcript compaction.
Caliber ai-setup: optional Jev transcript compaction → -
Reindeer-AI/pi-jev-guard
For the Pi coding agent: every code edit is checked by Jev against the repository's Markdown rules (agents.md and friends) so the agent stops ignoring them.
pi-jev-guard: check Pi code edits against repo rules → -
cclank/jevclip
Jev scores prepared subtitle or transcript segments; the tool assembles highlight and shortened videos with ffmpeg and exports a timestamped selection report.
jevclip: transcript-driven video highlights → -
cappuch/openjev.cpp
A llama.cpp fork serving Jev-like models as an inference server.
openjev.cpp: a llama.cpp fork for Jev-like inference → -
bespokelabsai/nimble
Bespoke Labs' open take on a Jev-like model: local typed decisions plus a contrastive data-curation recipe and evaluation code, released as data, weights and training recipe together.
Bespoke Nimble: open data, open model, open recipe → -
TrainLCD/Functions
A pull request on the TrainLCD Functions repository adds Jev-based re-ranking of suggested stations and measured spikes; the author says Jev is also going into part of the app's AI chat.
TrainLCD: Jev re-ranking in a production train app → -
mmastrac/djev-spark
A container recipe for structured decisions with DiffusionGemma in NVFP4 on an NVIDIA DGX Spark, offered as the 'or just use this' route to a local Jev-like.
djev-spark: DiffusionGemma NVFP4 decisions on a DGX Spark → -
mandu5/jevcompat
A testable System One specification and CLI check Jev-compatible servers and provide a normalizing proxy and fault-injectable reference mock.
jevcompat: System One API conformance suite → -
luantak/is-malicious
A codebase scanner that checks third-party repositories and CI scripts for malicious behaviour before execution, with Jev as the judge.
is-malicious: scan a repo before you run it → -
i-priyanshuverma/laravel-jev
Runtime `if (Jev::is($body, 'spam'))` and `Jev::choose($msg, [...])` for Laravel, with `Jev::fake()` for offline tests and `Rule::jev()` for form validation.
laravel-jev: Jev::is / Jev::choose inside Laravel → -
buberlo/dsh-jev
A DeepSeek Harness plugin uses Jev for tool relevance, pre-execution call assessment, skill hints and model-route selection through a reusable decision core.
dsh-jev: Jev decision layer for DeepSeek Harness → -
TheoLeeCJ/SemIf-OpenJev
An independent local probability-decision implementation inspired by Jev.
Jev-like model on a home GPU → -
danielgshea/jev-as-a-judge
LangChain tested Jev against LLM judges on accuracy, repeatability, latency and cost to see whether System One models offer a new approach to online agent evaluation; evaluator code in the repository.
Jev as a Judge: decision model vs LLM judges for evals → -
4esv/jev-eval
An independent comparison on Banking77, SST-5 and IMDB covering accuracy, calibration, latency and cost.
Jev vs GPT-5.6 Terra head-to-head → -
Anil-matcha/awesome-jev-by-typesafe
A large collection of use cases, patterns, prompts and starter code; despite the name it is a community repository.
Awesome Jev by TypeSafe (unofficial) → -
agent-chaperone/agent-chaperone
An MCP proxy and client hooks screen agent tool calls and results using Jev among other checks.
Agent Chaperone: MCP tool screening → -
jamesward/zio-typesafe-ai
A Scala ZIO client for Jev with a small DSL for typed questions.
zio-typesafe-ai: a Scala ZIO client → -
gholtzap/jev-codex-model-and-effort-router
An autorouter that lets Jev pick both the model and the reasoning effort for each Codex turn — 'I cannot use codex the same anymore'.
A Jev-based model and effort router for Codex → -
mattn/go-jev
A Go SDK pairs typed Jev requests with a CLI for line-by-line classification, semantic grep and JSON decision output in shell pipelines.
go-jev: Go client and UNIX decision pipelines → -
RenaGao/jev-dataops
A workflow for uploading, screening, training and evaluating datasets, with Jev API calls used for the screening stage.
JEV DataOps: screen and evaluate training data → -
maxlibin/jev-rubiks
An interactive 3D Rubik's cube: turn it by hand, Jev watches each move, stays quiet while you progress, warns on a break and steps in with the next algorithm.
jev-rubiks: the code solves, Jev coaches your hand solve → -
syumai/jevyoumean
Wraps commands and uses Jev to match mistyped subcommands by intent, not edit distance: `git record` → `git commit`, `npm add` → `npm install`.
jevyoumean: semantic 'Did you mean?' for any CLI → -
jagenaujagenau/ground-truth
A browser extension (and web version) that classifies the news article you are reading by political framing, article type, topic and how loaded the language is, using Jev as a fast probabilistic classifier.
Ground Truth: news-framing browser extension → -
mpuig/system-one
An independent Jev-inspired model and serving project for Apple Silicon. Fine-tuned open backbones return Choice, Score and Noul probabilities through a Jev-shaped API, with calibration artifacts and worked use cases.
system-one: calibrated local models on Apple MLX → -
qiyangdev/hn-jev
A Chrome extension scores Hacker News stories on five factors with Jev and reorders the page using adjustable weights.
HN Jev Scores: adjustable Hacker News ranking → -
Argos1111/jev_local
A TypeSafe-compatible adapter running LFM2.5-1.2B through llama.cpp (about 1.25 GB download) with CUDA, ROCm and Metal builds.
jev_local: TypeSafe-compatible adapter over LFM2.5 → -
qkal/Canny
Deterministic hooks decide, Jev advises: an append-only ledger with zero runtime dependencies that blocks an agent's 'done' until there is evidence.
Canny: stop coding agents from claiming work is done without evidence → -
Jabbslad/pi-jev-tools
A Pi extension exposes Jev ranking, classification, local snippet retrieval and general typed evaluation while retaining the normal coding model.
pi-jev-tools: ranking and retrieval tools for Pi → -
PPRAMANIK62/yc-jev-bench
An independent reranking benchmark paired with a live natural-language search over Y Combinator companies. The app uses Jev for query routing and reranking after keyword and embedding retrieval.
YC × Jev: startup search and a reranking benchmark → -
mustafakendiguzel/aso-workbench
An open-source Mac tool for App Store Optimization with keyword analysis and listing workflows that the author connects to Jev.
ASO Workbench: Jev-assisted keyword analysis → -
Teylersf/WindowsJev
A local MCP server for Windows that combines UI Automation actions with checkpointed web-research jobs. Jev selects from a bounded desktop action list and can rerank stored evidence or check a claim against an exact passage.
WindowsJev: bounded Windows actions and durable research → -
oboroge0/jev-voice-games
Two voice-controlled Godot games use local speech recognition for clear commands and Jev with game state for ambiguous ones.
jev-voice-games: voice-driven Godot mini-games → -
CommandCodeAI/cmd-mod-jev-nudge
A Command Code mod asks Jev whether a gentle nudge would help an agent continue useful work within the user's existing request.
cmd-mod-jev-nudge: decide when an agent should continue → -
mourad-ghafiri/qwen3.5_0.8B_decision_model
An independent 0.8B Qwen3.5 model for typed decisions, published with weights and a comparison against Jev.
Qwen3.5 0.8B decision model: a local Jev alternative → -
molis-ai/jev-workbench
A local browser workbench for defining, previewing and publishing versioned Jev functions that callers can invoke over HTTP or MCP.
Jev Workbench: versioned judgment functions → -
forestwas/gmail-jev
Classifies Gmail threads into workflow labels with Jev and can optionally remove them from the inbox. The tool does not send mail.
gmail-jev: label Gmail threads with Jev → -
Bhuwan-web/intent-classification
A modular gateway: define each agent's scope and confidence threshold, plug it in before the workflow — one reusable guard, many scopes.
intent-classification: a scope guard in front of agent workflows → -
Arpit-Khandelwal/jev-linkedin-slop-filter
A Chrome extension that applies BAIT, CORP and BRAG labels to LinkedIn posts using Jev, with a local server for the API key and heuristics to reduce calls.
jev-linkedin-slop-filter: label LinkedIn posts → -
typesafe-ai/system-one-adapter-python
A drop-in replacement for the TypeSafe client that answers the same state-plus-questions contract with an ordinary LLM API, useful for comparing Jev against OpenAI- or Anthropic-compatible models on identical calls.
System One adapter for LLM APIs → -
southleft/component-charades
A Taboo-style game where a player describes a UI component without its forbidden words and Jev ranks the component library. A second mode finds a component from the job the player needs it to perform.
Component Charades: Jev guesses a UI component → -
mejiasd3v/pi-jev-router
Open-source Pi extension that lets Jev route each turn to a model through Vercel AI Gateway; the author pairs it with Astra adaptive thinking.
pi-jev-router: automatic model routing for Pi via Vercel AI Gateway → -
JohGirard/doom-ai-overlord
Doom played by the Laya System-1 model via ViZDoom: one forward pass per decision, fully local, live model probabilities on screen; the model advises, deterministic rails enforce.
doom-ai-overlord: Doom on a 421M local decision model → -
gbesse/jev-lab
Fifty-nine small open-source projects exploring what changes when AI returns a typed decision instead of a paragraph — agent safety, RAG, research, games, public data — mapped in one repository.
jev-lab: 59 open-source projects around one question → -
SkywalkerDarren/feed-lens
An open-source Chrome extension (store review pending) that labels posts on X, Threads and Weibo with rules you define, powered by your own TypeSafe account.
Feed Lens: your labels on X, Threads and Weibo → -
sawzhang/jev-demo
A Chinese-language learning repository with five runnable Python demos and the author's Jev 1.13.0 experiments on question fan-out and decision boundaries.
sawzhang/jev-demo: five demos and Jev measurements → -
torumitsutake/jev-page-verdict
A Chrome extension that uses Jev to estimate a page's purpose and the writer's position (ad, testimonial, explainer…) and colour-codes Google search results accordingly.
Page Verdict: classify web pages by purpose and stance → -
0xNatoshi/jev-codex-router
Per-turn routing chooses a model, reasoning depth and speed mode for Codex.
Jev Codex Router → -
denser-org/rerank-bench-jev
An inspectable comparison of Jev Noul and Qwen3-Reranker-0.6B on retrieval ranking tasks, with the harness and raw model responses published.
rerank-bench-jev: Jev versus Qwen3 Reranker → -
jxspam/jev-ui-map
Maps visible UI states to bounded action options so Jev can select the next browser step. The repository includes a food-ordering demonstration with action limits in code.
jev-ui-map: typed actions for browser tasks → -
reachjalil/jevlogs
Open-source log triage for OpenTelemetry: Jev scores each log line's signal so only what matters reaches the expensive LLM analysis; a write-up covers Jev vs a fine-tuned alternative.
jevlogs: score log signal before the expensive LLM pass → -
Sac-Y/Jev-cu
An open-source Codex skill: install it, set a Jev API key, and Codex drives desktop apps with Jev making the per-step calls — the author reports the biggest gains on tasks with frequent checks and tool calls.
Jev-cu: the fastest computer use inside Codex → -
DumoeDss/logjev
Adapts first-token logprobs from any OpenAI-compatible endpoint (cloud or local) into bounded Jev-style choices and scores; text, image or audio in when the backend exposes logprobs. Node + Python, with agent skills and a 2048 demo.
LogJev: Jev-style choices from any model's first-token logprobs → -
jeffonelson/jev-bigquery-cloudrun
A Cloud Run remote function lets BigQuery call Jev to classify support tickets and expose the resulting judgments to SQL queries.
Jev decisions in BigQuery through Cloud Run → -
savka777/jev-use
Say it and the Mac does it: the loop reads the accessibility tree, asks Jev which element to act on, acts, repeats — deliberately naive, no screenshots, no vision model.
jev-use: a Mac computer-use harness without a vision model → -
chris-wozniczek/jev-voice-control
A macOS menu-bar voice controller with on-device speech recognition and Jev decisions for commands and UI actions.
Jev Voice: typed decisions for Mac voice control → -
thehumanworks/jevgrep
A standalone Rust binary searches a codebase using natural-language questions and returns relevant files and line ranges with scoring details.
jevgrep jg: Rust natural-language code search → -
bytelabs-oss/clash-jev
A Clash Royale experiment on an Android device that converts screen video into structured game state. Jev selects a strategy, card and placement, while the replay viewer exposes the questions and option probabilities.
clash-jev: live Clash Royale decisions and recorded replays → -
Jevals/jevals-data
An independent evaluation site and data repository for Jev and LLMs answering the same typed questions. Public boards link accuracy, calibration, costs and latency to human-labelled tasks and individual decision logs.
Jevals: labelled benchmarks and per-decision logs → -
zadescoxp/kadeconsole
A Bloomberg-terminal-style analytical tool in the terminal (`pip install kadeconsole`) with Jev in the decision path.
kadeconsole: a terminal-first stock analytics tool → -
iapp-technology/openthai-systemone
Five days after launch: a Thai + English System One model on Qwen3.5-0.8B with a 256-way slot head, Apache-2.0, sized to run on a laptop.
OpenThai-SystemOne: an open Thai + English decision model → -
morinokami/hunk-triage
An extension for bentlegen's hunk that classifies the files in a diff (core, tests, docs…) with Jev and reorders the review so the important changes come first.
hunk-triage: read the important changes first → -
chopratejas/invalidate
A layer that decides when an agent's stored memory should be invalidated, using Jev checks; a playground is hosted.
invalidate: memory-lease invalidation with Jev → -
zeredy879/minojev
An independent Jev-style project that freezes a language-model backbone and trains a small decision head over cached candidate features. It publishes typed probability outputs, datasets and browser presentations of its benchmark and game policies.
minojev: a trained decision head on a frozen language model → -
intikhab49/open-jev-typed-decision-engine
An open reproduction that answers noul/choice/score in one non-autoregressive pass with calibrated confidence, at 150M parameters — a non-LLM route to the same interface.
open-jev: a 150M non-LLM typed decision engine → -
kataras/jev
A community Go client for Jev's Noul, Choice and Score questions with typed response helpers, configurable retries and rate limiting.
kataras/jev: a Go client for System One → -
beamnxw/minelog
The harness behind minelog.xyz: GPT-6 Astra plans, Jev makes the moment-to-moment decisions for a single Minecraft agent.
MineLog: Astra plans, Jev decides, one Minecraft body → -
aowang-ai/jev-trade
A trading bot that reads the Hyperliquid book, asks Jev for long or short and open, close or hold, then places orders through separate execution code.
Jev Trade: Hyperliquid trading bot → -
Hu-xiao-max/jev_robot
Jev-inspired sim with a local 2B model driving a PiPER robot in MuJoCo: the cube is relocated three times per run and the robot detects, interrupts and recovers.
jev_robot: a moving-target pick-and-place in MuJoCo → -
VGabriel45/polymarket-btc5m-jev-trading
A Polymarket trading agent with a TUI where Jev is the decision layer for the 5-minute BTC Up/Down market.
Polymarket 5-minute BTC Up/Down agent on Jev → -
hr98w/jev-visual
An educational experiment: one image, dozens of questions answered at once by scoring candidate outputs directly instead of generating, running Qwen3.5-0.8B on MLX.
jev-visual: Jev-like visual inference on Apple Silicon → -
TianyuCodings/NanoJev
A 0.6B Qwen-based open implementation of parallel typed decisions, with training and evaluation code.
NanoJev → -
moritzkremb/jev-voice-browser
A local Node and Playwright app streams partial speech transcripts, asks Jev for intent and target choices, then applies code-level action gates.
Voice Browser: spoken commands to bounded browser actions → -
Dimweaker/jev-libero
A configurable LIBERO robot-control experiment that combines Jev decisions with reversible physics previews. An interactive replay follows intent, motion family and control choices alongside saved trajectories.
Jev × LIBERO: robot decisions with local physics previews → -
hotchpotch/jev-reranker
A Python package for Jev-powered relevance filtering and reranking, with a Hugging Face blog walkthrough covering the API, filtering and evaluation results.
jev-reranker: relevance filtering and reranking for RAG → -
lib-x/typesafe-go
Go SDK for the TypeSafe API: typed questions in, structured answers out, no dependencies.
typesafe-go: a zero-dependency Go SDK → -
mizzlelover/jev-hub
A Chinese aggregator of X long-form posts and demo videos about Jev, categorised with original links and authors kept; GitHub Pages site plus repository.
JEV HUB: 216 demo videos and 214 long posts, categorised → -
DomMonte/n8n-nodes-typesafe-ai
An n8n node with Ask Yes/No, Ask Choice, Ask Score, batch Evaluate and model lookup operations, defaulting to `jev-latest`.
n8n community node for the System One API → -
FogMoe/necro
LoRA fine-tuning experiments for Jev-like typed judgments on Qwen3.5-0.8B, published with datasets, adapters, evaluations and a full write-up of why it was abandoned.
necro: an abandoned Qwen3.5-0.8B LoRA replica, with the retrospective → -
madeye/pi-jev
Speeds up Pi coding-agent workflows by letting a decision model pick which files matter and cache requests; the author also tried DiffusionGemma in place of Jev and documents the results in VALIDATION.md.
pi-jev: Jev-assisted file retrieval and request caching for Pi → -
m-mizutani/semgate
Go HTTP middleware that lets Jev judge and route requests, plus a CTF-style example site inviting people to try injection attacks against it.
semgate: semantic request filtering for Go HTTP → -
pedroknigge/mcp_jev
A local MCP server that runs Jev "packs" (code audit, PR review, GUI automation) for Cursor, Claude Code and Codex.
mcp_jev: local MCP server with Jev packs → -
sutro-sh/jev-align
An open-source CLI that builds calibrated classifiers from human feedback on top of Jev, using GEPA prompt optimisation — because a foundation model still needs aligning to your own criteria.
jev-align: calibrate Jev to your decision criteria with GEPA → -
jayanthchandra/jgrep
A Rust CLI that pipes logs, files or any text through Jev: natural-language question in, structured semantic decision out.
jgrep: ask questions of logs, files and pages from the shell → -
wdwxw/jevdemo
A small Python and HTTP example of support-request routing with Choice, frustration scoring with Score, and urgency evaluation with Noul.
jevdemo: Choice, Score and Noul in Python → -
larguesa/jev-search
A Python command-line tool uses Jev to find relevant lines in selected local documents and knowledge-base text alongside lexical search.
larguesa jev-search: semantic line retrieval → -
Heman10x-NGU/openJev-verdict-2.0
A non-autoregressive decision engine on ModernBERT (151M) with calibrated uncertainty, an audit of the Jev benchmark, and an in-browser WebGPU playground; Apache-2.0 with PyTorch and ONNX weights on Hugging Face.
openJev-verdict 2.0: a 151M calibrated decision engine → -
Peu77/JevFind
A Rust code-search CLI first scores source paths with Jev, then ranks overlapping windows from the selected files and prints matching snippets.
JevFind: two-stage semantic code finder → -
0xtrou/rubikjev
Scramble a Rubik's cube and watch Jev judge and solve it live, with difficulty tiers, XP and token-metered runs; labelled experimental and fan-made.
Jev solves the cube → -
ChenReuven/xo-jev
A browser tic-tac-toe game where Jev chooses the O move from the currently free cells.
xo-jev: Jev plays tic-tac-toe → -
mizchi/jev-lexer
Split like gpu-lexer, classify every token with Jev, render Shiki-compatible tokens to HTML and ANSI — a language-agnostic highlighter, experimental only.
jev-lexer: syntax highlighting with no language knowledge → -
muse0509/jev-preflight
Born from Claude Code changing an auth check to return true with tests still passing: eight risk axes scored by Jev in one request, with one optional re-inspection before the agent declares done.
jev-preflight: a bounded risk check for Claude Code → -
rmalde/minecraft-agent
A Minecraft agent that plans with Astra and lets Jev control step-by-step actions, with native recording, tested routes and run verification; the Jev decision rules live in models.mjs.
minecraft-agent: Astra planner, Jev controller → -
choxos/jevchess
A one-page web app where Jev plays chess against any OpenRouter LLM, Stockfish, or a human, showing live moves, Jev's move probabilities, saved games and win rates.
JevChess: Jev vs LLMs, Stockfish or you → -
logan-markewich/jeff
A drop-in for TypeSafe's API powered by a 400M GLiFormer model; point the official SDK's base URL at it. Deploys on Modal or CPU via ONNX Runtime and ships a bench folder.
jeff: a self-hosted Jev API on GLiFormer → -
lukeramsden/system-one
A TypeScript client with shared typed question definitions and Promise and Effect APIs for Jev and other System One-compatible models.
system-one: Promise and Effect clients for typed decisions → -
CorieW/JevTest
A local TypeScript library and CLI where Jev chooses exploratory browser actions and assesses observed results. Playwright executes configured actions, exact assertions decide outcomes, and saved traces support replay.
JevTest: exploratory browser tests with replayable evidence → -
enderkus/zammad-jev-dispatcher
New Zammad ticket comes in, Jev decides which team owns it with a confidence score attached, and routes it.
Zammad ticket dispatcher on Jev → -
devagrawal09/stanley-code
Bounded TypeSafe Jev workflows for coding agents: small-model judgments combined with deterministic workflows so coding work stays bounded, from the author of jev-review.
Stanley: a Jev-first, self-evolving coding agent → -
TenzinDhonyoe/bev
`npx bev-ai`: an open-source joke-but-working classifier with bring-your-own Jev key and pluggable status lines.
bev: the slowest classifier in the world, at Jev prices → -
krzyzanowskim/TypeSafe
A second Swift SDK, from the author of CryptoSwift — 'just leaving it for Jev enthusiasts'.
TypeSafe SDK in Swift (krzyzanowskim) → -
MANISH007700/tab-bouncer
Tell the extension what you are working on; Jev checks every open tab at the door in a single call and shows the freeloaders out.
Tab Bouncer: close the tabs you don't need, judged in one call → -
milind-soni/tiptour-macos
An open-source macOS pointer companion driven by Jev or Gemini Live: a local OmniParser build on CoreML reads the screen, Jev picks where the pointer goes, Ctrl+K takes a typed task. Bring your own keys.
TipTour: a local pointer companion for macOS → -
AdamPippert/granite-decisions
A Jev-style decision model built on IBM Granite 4.1 3B with an RHCD training recipe — 'when's IBM or Red Hat going to build their own Jev? I got it for them.'
granite-decisions: a Granite 4.1 3B RHCD model → -
shunshunNi/redmine_jev
A Redmine plugin that asks Jev, while you fill in the form, whether the chosen tracker matches the title and description, using the project's recent trackers as options.
redmine_jev: does this issue's tracker fit? → -
marc2332/findme
'jev, find me stuff in my system' — a file finder that asks Jev which files match a natural-language request.
findme: jev + fs → -
achimala/jev-paint
A terminal art experiment where Jev makes the pixel decisions; renamed out of respect for an earlier project with the same name.
jev-paint: use Jev to make art → -
TheEleventhAvatar/triage-bot
Jev routes each ticket to a specialist agent (general / account / billing / technical) and decides whether a human should take it instead — all typed — while Cerebras writes the reply.
triage-bot: real-time support triage with Jev and Cerebras → -
kylemclaren/jevql
Adds `jev()` semantic predicates to plain Postgres through a CLI, HTTP and SDKs, with a SQLite cache and an `--explain` cost preview; a playground is hosted.
jevql: semantic predicates for vanilla Postgres → -
gargpratyush/jev-router
Routes Claude Code tasks to a cheaper or more suitable model based on a Jev decision; listed in the Studio Yebisu roundup.
jev-router for Claude Code → -
sensahin/transcript-lens
Browse YouTube transcripts by meaning with Jev analysis, subtitle export and a Vercel setup guide; Turkish UI.
transcript-lens: explore YouTube transcripts by meaning → -
ielab/llm-rankers
The ielab document-ranking toolkit gained a jev module, with code that uses TREC human relevance judgments to test whether Jev's probabilities are actually calibrated.
llm-rankers: checking Jev's calibration with TREC judgments → -
ajensenwaud/hermes-jev-plugin
A small plugin exposing Jev to Nous Research's Hermes Agent as tools: jev_check, jev_route, jev_score and jev_evaluate.
hermes-jev-plugin: decision tools for Hermes Agent → -
iammrduncan/typesafe-ai-benchmark
An LLM gateway that mimics TypeSafe's structured output so local models can be scored against Jev; Needle 3 on an M4 Pro came out worse than Jev, Qwen3.8 27B and OSS 120B.
Needle 3 vs Jev: an 'imposter Jev' gateway benchmark → -
skeptrunedev/jev-recruiter
A Jev-driven recruiting agent with a local web UI that shows evidence quotes behind each decision; a short demo video is linked.
jev-recruiter: LinkedIn recruiting agent → -
ranjan2829/AskJev
A Chinese X Article sorting the early ecosystem into things that actually ship: jev-ultrafast, mobile-jev, typesafe-computer-use, AskJev, a Twitter timeline analyzer, a tax-document classifier, jevdetector and newsjack, plus the OpenRouter and DataCamp explainers.
Got Jev — now what? A real deployment list → -
gopalanj/chess-jev
Not Jev but Jev-shaped: a small decision model trained specifically for chess moves, built in the launch-week spirit.
ChessJev: a System One model for chess → -
matthewp/flue-jev-demo
A Cloudflare Worker demo where a Flue agent uses Jev through Cloudflare AI Gateway for intent and urgency routing.
Flue agent + Jev via Cloudflare AI Gateway → -
githubnext/localjev
GitHub Next's 'poor man's Jev': a Jev-compatible local API built in a morning on top of DiffusionGemma (swappable via config), so team members without Jev access can run the same call shapes locally.
LocalJev by GitHub Next: a Jev-compatible local API → -
kbhuw/jev-sift
A portable agent plugin and MCP tool for batch text classification: the agent asks Jev whether a file is worth reading, a tool worth calling, a website worth browsing — before spending tokens on it.
jev-sift: classify first, read selectively → -
realZachi/pg-jev
A PostgreSQL extension that exposes Jev-backed categorical judgments over table rows.
pg-jev → -
Skyvern-AI/jevscape
Adapts the RuneBench RuneScape agent benchmark to Jev with about fifty bounded actions, a tick-mode controller and a live dashboard.
jevscape: RuneScape benchmark harness for Jev → -
zostaff/agent-arena
A 'bouncer called Jev' added to an agent trading arena that blocks bad trades before the AI even sees them.
agent-arena: a Jev bouncer for Degen Village → -
pithings/advocaat
A TypeScript client for batching several typed questions about your data into one Jev request; the example triages GitHub issues.
advocaat: a small type-safe client → -
obie/ruby_decision_model
A zero-dependency Ruby client for decision models supporting TypeSafe's native API and OpenRouter, with retries, timeouts and usage/cost fields.
ruby_decision_model gem → -
h100envy/nerve
A trading agent guard where Jev vets trades before the LLM commits; the author says it stopped 3 ETH going into dead pools in the first five days.
nerve: a Jev gate in front of an onchain trading agent → -
rimusz/jev-mcp
A proof-of-concept MCP server exposing three tools: verify a claim against evidence, screen text for injection or relevance, and rank candidates semantically.
jev-mcp → -
markfive-proto/typesafe-vs-deepseek
A side-by-side of Jev and DeepSeek-flash on invoice extraction, email classification and reranking, with numbers on a hosted dashboard.
Jev vs DeepSeek-flash side by side → -
DanRWilloughby/snifftest
Countable rules plus one judgment model: ten yes/no questions per paragraph flag AI-writing tells. Zero dependencies.
Sniff Test: a prose linter where Jev is the judge → -
FirasSX914/Janus
Measures on Banking77 when Jev can answer alone versus escalate to DeepSeek V4-Pro by confidence threshold, with risk-coverage calibration analysis.
Janus (formerly calibre): confidence-routed fallback study → -
r-ms/mini-jev
An X Article roundup of early projects: jev-ultrafast, jevlike, mini-jev, reflex, awesome-jev-by-typesafe, the Every vibe-check test and several video demos, with the official docs and use-case map linked.
WTF is Jev? Nine things people are already building → -
0xArx/jevegis
A bring-your-own-key API with /v1/scan (prompt injection, jailbreak, credential leaks) and /v1/moderate (nine categories) returning calibrated probabilities; free tier live.
Jevegis: guardrails API on Jev → -
jexp/neo4jev
Outgoing relationships become choices, Jev returns next-hop probabilities, and a beam search walks the graph; shipped as a library, notebooks and a Streamlit app on Neo4j's public Companies knowledge graph.
neo4jev: graph navigation with Jev → -
zhuyansen/jev-search-rerank-eval
Evaluates Jev reranking against bge-m3 embedding search over an Agent Skills Hub catalog, with judge-bias and robustness analysis.
Jev as a reranker: graded-relevance eval → -
yusukebe/hono-jev-router
An experimental semantic router for Hono by its creator: routes are described in plain language and Jev matches incoming requests to them from method, URL, headers and body.
hono-jev-router → -
colliber/duckdb-jev
A DuckDB extension with jev_choice(), jev_score(), jev_noul() and jev_ask() returning real SQL types, batching 16 rows per request and caching identical queries.
duckdb-jev → -
kitze/unclutter
A WXT browser extension that asks Jev whether page elements are clutter (ads, cookie banners, upsells), then saves reusable per-template hiding rules; sends snippets rather than full pages, bring your own key.
Unclutter: Jev-powered clutter blocker → -
mrnugget/jev-shell-history
Picks the most likely command from the last 100 unique history entries and ghosts it after the cursor, fish-style, using Jev to rank by intent.
jev-shell-history: fish-style zsh suggestions ranked by Jev → -
teyhouse/jev-secret-detection
A quick credential and secret-detection experiment with Jev; the author reports consistent results across runs.
Secret detection test with Jev → -
tamaratran/fast-jev-compaction
Jev scores which tool calls and results remain relevant before context is compacted.
Instant context compaction → -
akash-kamat/system-one-gemma
Reimplements a System One decision model on Gemma 3 270M with a LoRA of 2.6M trainable parameters, scoring all options in one forward pass; trains in a free Colab and ships weights in the repo.
system-one-gemma → -
emrickgarrett/OneVOneJev
A browser 1v1 quickscope arena built with Three.js where Jev plays the opponent's decisions in real time.
OneVOneJev: a 1v1 quickscope arena on Three.js → -
trungdq88/youtube-sponsor-detection
An open Chrome extension listens to live audio and skips detected sponsor segments.
Live YouTube sponsor detection → -
sgoedecke/system-one
Constrains token selection to choice indices so any open LLM answers many questions in one pass; includes Doom and Wikipedia-racing demos on Qwen models.
system-one: batched single-token choice inference → -
phyous/tsai-civ2
An experimental browser harness for the original Civilization II that asks Jev to choose empire, city, research, diplomacy and unit actions. It records player-visible observations and model probabilities in a spectator display.
Jev × Civilization II: an experimental campaign harness → -
Stumble/jev-go
A community Go SDK implements Jev questions through TypeSafe and Vercel transports and includes an interactive command-line wizard.
Stumble jev-go: Go SDK and interactive CLI → -
TokenTrim/jev-routing-experiment
Benchmarks Jev as a router across 13 flagship models on LLMRouterBench and a RouterArena pilot.
Jev as an LLM router: a mostly negative result → -
standardagents/jevpilot
A playable Three.js driving simulation where Jev selects sampled steering and speed paths from compact road, traffic and route state.
JevPilot: browser driving simulator → -
dbreunig/building-with-jev-skill
An agent skill for finding and designing bounded Jev decisions in software.
Building with Jev skill → -
JacobLinCool/jev-paper-judge
Upload a research-paper PDF; text is extracted in the browser, split into roughly 1,000-word passages, and a fixed set of small questions is asked per passage in parallel, then combined into weighted scores.
Jev Paper Judge → -
caiovicentino/jev-shield
Sits between MCP clients and servers with three layers: structural allowlists, semantic hazard detection via Jev, and session taint tracking; targets poisoned tool descriptions, injection, exfiltration and secrets.
jev-shield: semantic MCP firewall → -
Wizhill05/typesafe-image-diffusion
Turns Jev into a pixel-art generator by asking 256 multiple-choice questions, one per pixel, plus refinement passes.
Jev as a pixel-art generator → -
altryne/twitter-timeline-analyzer
A Chinese X Article sorting the early ecosystem into things that actually ship: jev-ultrafast, mobile-jev, typesafe-computer-use, AskJev, a Twitter timeline analyzer, a tax-document classifier, jevdetector and newsjack, plus the OpenRouter and DataCamp explainers.
Got Jev — now what? A real deployment list → -
0x963D/last-exit
A text game where you talk your way past a border inspector whose actions are chosen by Jev from your typed replies; dialogue and consequences are authored game code, Jev only decides.
Last Exit: a checkpoint game judged by Jev → -
zsoXi/agent-handoff-gate
An experimental protocol where a worker agent sends requirements, code and test results to Jev, which returns an evidence verdict for the team-lead agent before results are accepted.
Agent Handoff Gate → -
HackSing/jev-report
A 52-page report with 50 reproducible Chinese test cases and raw API responses that re-measures TypeSafe's headline speed and cost multiples.
Chinese-language fact-check of the 193.6x / 444.6x claims → -
droidrun/mobile-jev
A mobile agent that drives a real Android device through Mobilerun with Jev choosing actions; the README states the limits of the demo itself.
mobile-jev: an Android agent on Jev → -
ChetasLua/jevmeter
An open-source tool that fact-checks spoken content (debates, investor calls, interviews, podcasts) live with Jev calls; demoed on an interview.
jevmeter: live fact-checking of spoken content → -
adtyavrdhn/pydantic-jev-examples
Pydantic AI merged a TypeSafe model provider: `TypeSafeModel('jev-latest')` with confidence thresholds, a FileUnderstanding capability and fallback to an LLM for tool calls; docs cover jev-latest, jev-preview and pinned releases.
Pydantic AI: first-class TypeSafeModel → -
mateonunez/jod
Validates state locally with Zod-style schemas, bundles all questions into one parallel Jev request and projects typed answers; built on the official SDK.
jod: semantic schemas over Jev → -
rorshopping/jev-on-a-laptop
Prefill once, broadcast the KV cache per field, evaluate all fields in one pass on Qwen 1.5B–8B models; a Hugging Face Space demo exists.
jev-on-a-laptop: parallel constrained decoding on MLX → -
AbdelStark/jev-benchmarks
A framework comparing hosted Jev with a local GLiNER2.5 on zero-shot single-label classification (AG News, Banking77, DAIR Emotion) for accuracy, calibration, selective-automation coverage and latency, with bootstrap intervals.
Jev vs GLiNER2.5 benchmark → -
mithalouni/system-one-open
An open replica trained on 92 public datasets plus synthetic generators, on Gemma 4 E2B (LoRA) and Gemma 3 270M, served via FastAPI on Modal with a live demo.
system-one-open → -
legacybridge-tech/pi-typesafe-jev
A Pi extension exposing TypeSafe judgments as five tools (choice / bool / score and friends) so the model makes narrow semantic calls while your code keeps thresholds and weights.
pi-typesafe-jev: Jev as five native Pi tools → -
jarrodwatts/jev-trader
A bot that asks Jev for a buy/sell decision on every Monad block from a price feed and executes on Kuru's MON/USDC book, with a live status page.
Jev Trader: one decision per Monad block → -
prismhq/jev-router
An OpenAI-compatible router that uses Jev to pick a model on top of LiteLLM, filtering candidates by capability and falling back to the cheapest model without a TypeSafe key.
jev-router on LiteLLM → -
devagrawal09/jev-review
A step-by-step code-review workflow built on Jev with a local dashboard, listed in the Studio Yebisu roundup.
jev-review: staged code review with a local panel → -
JoshuaSP/open-jev
Uses an open diffusion language model for structured JSON judgments without fine-tuning and benchmarks against Jev's public eval cases.
open-jev with DiffusionGemma → -
GiesN/typesafe-jev-workflow
An async LangGraph workflow that classifies mock emails (invoice versus general) with the Choice primitive and routes them to handlers, with unit tests; pins jev-1.12.
LangGraph email workflow with Jev → -
vinnylarouge/jevlike
An early independent implementation of a Jev-style interface on an open model; cited as inspiration by daseinlabs/open-jev.
jevlike → -
sharziki/semdecide
Jev as a command-line filter: classify, score and filter lines in a shell pipeline, for crawlers, CI gates and data pipelines.
SemDecide: typed semantic decisions for Unix pipelines and CI → -
ellipsis-dev/blink
Jev scores file and folder names so an ensemble of walkers is allocated across the most probable paths; a Bun CLI with recursive and multi-walker modes and JSON traces.
blink: codebase search with Jev-scored paths → -
Twister915/typesafe-ai
A Rust crate with enum responses and confidence, async or blocking backends, batched questions and observable retries; examples cover PR questions and semantic search.
typesafe-ai: Rust bindings for the System One API → -
phyous/tsai-sc
A harness for the original StarCraft shareware campaign that gives Jev structured visible game state and executes its choices through keyboard and mouse inputs. The release includes a probability-overlay video and an inspectable run bundle.
Jev plays StarCraft: a recorded Strongarm mission → -
fhshaik/typesafe-mario
Jev picks NES controller inputs from object-centric JSON telemetry rather than screenshots, choosing among seven legal actions and logging latency and probabilities per decision.
Jev plays Super Mario Bros from telemetry → -
typesafe-ai/typesafe-sdk-js
The official TypeScript/JavaScript client, shipping ESM, CJS and type declarations for Node 20+; same ticket-classification quickstart as the Python SDK.
@typesafe-ai/sdk (TypeScript) → -
typesafe-ai/skills
TypeSafe's public skills repository with the `typesafe-ai` SKILL.md used by Claude Code, Codex and other agent environments to design bounded Jev decisions.
Official agent skills repository →
No repository activity to show
Try a broader search or .
Last push records repository activity across branches and tags. Release announcements and descriptions of individual changes need their own sources. Dates are shown in UTC; “Metadata checked” belongs to that repository alone.