Research & evaluation
Find research and evaluation projects exploring decision models and Jev-related approaches.
- 20jevlikeCommunity Jev-like model and tooling experiment inspired by TypeSafe System One.
vinnylarouge/jevlike · Python
1,371122model - jev-experiments: TypeSafe Jev ecosystem repository.
dabit3/jev-experiments · TypeScript
41334typescript - 64jevbenchJevBench v1 - a benchmark for Jev-class typed decision models: smart, cheap, fast, reliable, open.
fstandhartinger/jevbench · Python
34333python - Camera-only autonomous drone in MuJoCo with a small judgment model (TypeSafe Jev) in the loop at 2.5Hz
RomanSlack/jev-drone · Python
30041 - 111RSI-JevTyped-decision models (noul / choice / score) trained by a self-improving loop of AI agents — checkpoints, the code that produced them, and every version that failed.
Shanghua-Gao/RSI-Jev · HTML
13712ai-agentsautonomous-researchdecision-modelllmrecursive-self-improvementtyped-decisions - 123jevalsAgent evals and guardrails as Jev decisions: one request per trace, a fraction of a cent, fast enough for the agent loop. Runs locally with Kev or Laya.
openlayer-ai/jevals · Python
12013ai-agentsevaluationguardrailsllmllm-evaluationragas - Using Jev as an evaluator.
danielgshea/jev-as-a-judge · Python
11417python - 154jevfireJEV-inspired parallel decisions for CUDA LLMs. One context, many decisions. vLLM API, game-agent examples, and reproducible benchmarks.
kikoncuo/jevfire · JavaScript
779javascriptcudagame-aiinference - 170RoboJEVTwo-stage JEV control of a Franka Panda in MuJoCo
lykycy123/RoboJEV · Python
643python - 172jev-designCan a model design a dashboard? A console whose whole design system is generated at runtime by Jev from a one-sentence brief.
bilune/jev-design · TypeScript
634typescript - The highly anticipated open-source repository for JEV as Policy enables one-click setup of the simulation environment. Evaluations of Astra + JEV on benchmarks such as RoboTwin will also be released soon.
YuanKJing/Jev-as-Policy · Python
503python - Open reproduction of TypeSafe Jev: a 150M typed decision engine (noul/choice/score in one non-autoregressive pass, calibrated confidence). 0.697 vs Jev's 0.727, 2.5x better calibrated, 4x faster, free. Trains on a Colab T4 in 30 min.
intikhab49/open-jev-typed-decision-engine · Python
504agent-observabilitycalibrationcolabdecision-modelexpected-calibration-errorhuggingfacellm-alternative - 205open-jevTyped JSON inference with DiffusionGemma, with Every and Jev benchmark results
JoshuaSP/open-jev · Python
442python - Open replica of TypeSafe's Jev: typed calibrated decisions in one forward pass, on Gemma 4 E2B / Gemma 3 270M (Modal)
mithalouni/system-one-open · Python
417python - This is a LLM Gateway that mimics typesafe ai structured output. Like an imposter Jev.
iammrduncan/inference-benchmarks · TypeScript
395 - 230openjevLocal bilingual probability decisions from context, questions, and candidate answers. Independent research preview inspired by TypeSafe Jev.
zhihz/openjev · Python
374python - 235pix-golpeDemo: IA detecta golpe do Pix e rastreia a quadrilha. Rust + Jev vs Python + DeepSeek em tela dividida.
sandeco/pix-golpe · Python
3613python - 237jev-idsBlazing-Fast Token-Efficient Intrusion Detection System (IDS) based on TypeSafe's Jev.
jev-sec/jev-ids · Python
342cybersecurityintrusion-detectionintrusion-detection-systemnetwork-securitypython - Weekly builds on Jev (TypeSafe AI), benchmarked honestly enough to publish. Week 01: inbound lead triage, 90% routing at 366ms and $0.04 per thousand leads.
deepdave98/jev-playground · Python
333benchmarkgtmlead-scoringllmrevopshtml - Independent, evidence-based map of when TypeSafe's Jev actually holds up vs. breaks down — real API-call receipts, not a leaderboard. 中文為主的雙語 repo。
Zaious/jev-capability-atlas · Python
306ai-agentsbenchmarkcalibrationllm-evaluationmachine-learningzh-tw - Probability-aware evaluation for typed decision models: calibration, selective risk, latency, and reproducible benchmarks.
AbdelStark/jev-benchmarks · Python
285pythonbenchmarkingcalibrationevaluation - Unofficial study: Jev-style parallel typed decisions on stock 1.5B-8B models on an Apple Silicon laptop. Benchmarks, research notes, and a Hugging Face Space demo.
rorshopping/jev-on-a-laptop · Python
261python - 285jevalMeasures what your Jev classifier's confidence is really worth, and sets the human hand-off line from what a mistake costs.
rlaope/jeval · Python
252python - Real-time drone swarm autonomy simulation using Jev for fast System 1 reflex decisions and collision avoidance, with optional System 2 reasoning for strategic guidance
khordoo/jev-reflex-autonomy-lab · TypeScript
243autonomous-agentsdrone-simulationmulti-agentsystem-1system-2typescript - Jev vs Gemini 3.8 Flash: labelling 1,000 app reviews, 4.1× faster and 7× cheaper
goodrahstar/jev-column-race · JavaScript
233javascript - 323typed_evalsFast, typed, calibrated evaluations for LLM and agent outputs, powered by Jev — with simple, framework-agnostic Python APIs
TrustifAI/typed_evals · Python
191ai-agentscrewaievaluationevaluation-frameworkevaluation-metrics - 329jevscapeRuneBench harness for TypeSafe's Jev: bounded action catalog, tick-mode controller and a live dashboard
Skyvern-AI/jevscape · TypeScript
182typescript - 335jev_stockAn experimental JEV-powered framework for forecasting short-term stock price direction from structured market data.
sosopop/jev_stock · Python
175pythonfinancial-datafinancial-machine-learningllm - 343jev-dspy-labReproducible calibration and selective-risk benchmarks for Jev/TypeSafe decisions in DSPy workflows
jmanhype/jev-dspy-lab · Python
162python - 360jevbetterA stronger one-pass scorer over a variable list of text options. Hashed n-gram encoder, rival-aware attention, gated head, temperature scaling — with a head-to-head benchmark vs the jevlike starter design.
olanotolu/jevbetter · Python
153python - Interactive experiments with TypeSafe Jev, from support routing to 3D driving simulations with real AI decisions and visible sensor inputs.
kavehmz/typesafe-playground · JavaScript
154javascript - Jev vs DeepSeek 医疗任务对比:12 类场景、96 项测试,效果、耗时与费用一览。 | Jev vs DeepSeek on 96 healthcare tasks across 12 scenarios: quality, latency, cost and open experiment data.
JuneYaooo/jev-healthcare-lab · Python
131benchmarkchinese-nlpclinical-nlpclinical-trialsdatasetsdeepseekevidence-based-medicine - 397trade-jevBacktest Jev (TypeSafe) as a BUY/SELL/HOLD trader on NQ L10 order-book data
justinhe16/trade-jev · Python
133python - 401what-the-jevWhat can Jev actually do? Reproducible experiments and research reports exploring its capabilities and limits.
keta1930/what-the-jev · Python
131benchmarkdecision-modelevaluationllm-evaluation - Independent, reproducible benchmark: a decision model (jev), eight constrained LLMs, and deterministic baselines on typed decisions - accuracy, calibration, latency, cost, failure modes
nibzard/decision-model-benchmark · HTML
121benchmarkcalibrationdecision-modelllm-evaluationreproducibilityhtml - 410jev_fsdA driving simulator on real OpenStreetMap streets where TypeSafe's Jev model drives the car, with a decision inspector and a benchmark.
BrendanH18/jev_fsd · JavaScript
123benchmarkdriving-simulatorjavascriptopenstreetmappythonself-driving - 416JevNextJevNext · More than Choice — A simple algorithm that equips any Jev-like model with numerical control.
Bring-AI/JevNext · Python
123decision-treesinterval-decodingnumerical-reasoningreproducible-researchpython - Can a decision model beat dedicated rerankers? TypeSafe Jev vs Cohere Rerank 4 vs ZeroEntropy zerank-2 vs a chat-model baseline: 14 datasets, every raw API response, bootstrap ranges on every gap.
anessbelbati/jev-rerank-bench · Python
110python - 430JevLightJev-powered traffic signal control on CityFlow with structured phase and green-time decisions.
usail-hkust/JevLight · Python
113intelligent-transportationpython
1–50 of 389