A free GitHub project directory for the TypeSafe Jev / System One model.Explore official SDKs, agent tools, open-source alternatives and Jev news.

Data updated

All tags

Projects tagged with “calibration”

Browse Jev GitHub projects tagged with calibration. Filter, sort and explore their source repositories.

  • Independent open-source Jev-style decision model that turns open LLMs into typed Choice, Noul, and Score judgments with probabilities.

    nokia-applied-research/AnyJev · Python

    1,611180
    calibrationdecision-modelllmtransformersvllm
  • 1207 public resources for Jev, TypeSafe AI's System One decision model, indexed by decision pattern. Source citations, dated link checks and scheduled call-site text checks; runtime and performance are not independently tested here. EN/中文, JSON schema and platform compatibility.

    kydlikebtc/awesome-jev · Python

    60234
    agent-toolsai-agentsawesomeawesome-listcalibrationcatalogclassification
  • 🔥🔥 Papers, open reproductions and independent evaluations behind System One models and Jev.

    OmniJev/awesome-jev-gallery · Python

    49443
    awesomeawesome-listcalibrated-probabilitiescalibrationdecision-modelllmllm-evaluation
  • 🚀🚀 A multimodal System One decision model that gives calibrated answers to typed questions about screens, photos, video and text in one forward pass.

    OmniJev/OneJev · Python

    27328
    calibrationcomputer-usedecision-modelgui-agenthuggingfacellmmultimodal
  • JevK5: open-weight alternative to TypeSafe Jev. Typed decisions with probabilities in one forward pass; Apache-2.0 weights and code.

    allebee/jevk5 · Python

    16812
    calibrationdecision-modeljev-alternativejevbenchloraqwen
  • Unofficial skill that teaches coding agents to build with TypeSafe AI's Jev: typed decisions, calibrated confidence, and prior art from 150+ community projects.

    aaddrick/building-with-typesafe-jev · Python

    15612
    agent-skillsai-agentscalibrationclassificationclaude-codeclaude-code-plugin
  • Open-source alternative to TypeSafe's Jev: a System One style model layer that gives typed, calibrated decisions from any open-weights LLM in one forward pass (HF + vLLM), with honest benchmarks

    ikermoel/open-alternative-jev · Python

    8014
    calibrationclassificationjev-alternativellmlogprobsopen-jevopen-source-jev
  • Local decision engine with typed, calibrated outputs (Noul/Choice/Score): cheap→strong cascade, ECE/Brier calibration, SystemOne protocol interop. Apache-2.0.

    slavabars/decision-cascade · Python

    730
    calibrationcascadedecision-makingfastapillmmodel-routingpytorch
  • Curated list of tools related to system one models

    andyrewlee/awesome-system-one

    6215
    ai-agentsawesomeawesome-listcalibrationcomputer-useintent-classification
  • 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

    504
    agent-observabilitycalibrationcolabdecision-modelexpected-calibration-errorhuggingfacellm-alternative
  • 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

    306
    ai-agentsbenchmarkcalibrationllm-evaluationmachine-learningzh-tw
  • Probability-aware evaluation for typed decision models: calibration, selective risk, latency, and reproducible benchmarks.

    AbdelStark/jev-benchmarks · Python

    285
    pythonbenchmarkingcalibrationevaluation
  • Evidence-backed use cases, patterns, and guidance for building with Jev, TypeSafe AI's System One model. Every claim is labeled and sourced.

    aliaihub/awesome-jev-usecases

    247
    ai-engineeringawesome-listcalibrationstructured-output
  • Train JEV-style System 1 decision models on Qwen3.5 and Qwen3.8-27B: 24-slot decision head, KL distillation, calibration and serving. 在 Qwen 上训练 JEV 风格的 System 1 决策模型。

    shibing624/train-jev · Python

    212
    calibrationdecision-modeldistillationfine-tuningllmloranlp
  • Evidence-backed index of real-world Jev (TypeSafe AI System One) use cases: repos, patterns, benchmarks, and measured results

    anandi1989/awesome-jev-usecases

    195
    ai-agentsawesome-listbenchmarkbrowser-agentscalibration
  • openvons (open-Jev): 有限選択肢に確率で答える判断層 — テキスト / 画像 / 日本語音声コマンド

    genai-craft/openvons · Python

    150
    pythoncalibrationdecision-modeljapanese
  • 🌳 Groot 1 — open typed-decision model. Calibrated choice/score/yes-no answers over 64k-token states, System One API compatible, with a pure Bend engine. Beats Jev on zero-shot Folha news (87.8% vs 85.8%) and typed decisions (75.3% vs 73.7%).

    leofmarciano/groot · Python

    143
    bendbend-langcalibrationdecision-modelllm-routinglong-contextmachine-learning
  • A decision model that returns calibrated answers to yes/no, choice, score and multi-label questions.

    SamratDuttaOfficial/WaterSheep · Python

    144
    browsercalibrationdecision-modelhuggingfacejavascriptjev-alternativelocal-first
  • 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

    121
    benchmarkcalibrationdecision-modelllm-evaluationreproducibilityhtml
  • Three 4B decision models for English, Turkish and German, with per-option probabilities, GGUF builds and the Jev API.

    mertkayacs/jevalt · Python

    91
    calibrationdecision-modelenglishgermanggufjevaltllama-cpp
  • Tests, calibration audits and failure-mode studies of Jev (TypeSafe System One): jaggedness, consistency, injection, abstention.

    Yifan-Lan/awesome-jev-robustness

    84
    awesome-listcalibrationrobustness
  • Awesome Jev: an evidence survey of Jev and Jev-like typed decision models — calibration, selective control and open implementations, with a searchable literature site.

    Eurekaleo/awesome-jev-survey · HTML

    83
    awesome-listcalibrationevidence-synthesisselective-predictionsurveytyped-decision-modelshtml
  • Curated Jev and System One papers, models, datasets, benchmarks and SDKs, with versioned sources and transparent verification levels. English / 中文.

    reacher-z/Awesome-Jev · HTML

    71
    ai-agentsawesomeawesome-listbenchmarkcalibrationcurated-listdecision-model
  • Open-source Jev-style System One decision model. Gemma 3 270M with a scoring head — fast, calibrated decisions in a single forward pass. No text generation. Inspired by TypeSafe.ai's Jev.

    akash-kamat/system-one-gemma · Python

    72
    pythoncalibrationdecision-modelgemma
  • Unofficial skill for building with decision models (Jev, Clef, pplx-decider, ai_decide, Ollama): typed decisions, calibrated confidence, provider differences, and prior art sorted by implementation shape.

    aaddrick/building-with-decision-models · Python

    60
    agent-skillsai-agentsai-decidecalibrationclaude-codeclaude-code-pluginclef
  • First independent head-to-head benchmark of System One decision models (Laya vs Jev) on byte-identical inputs

    instax-dutta/sysone-bench · Python

    63
    benchmarkcalibrationdecision-modelevaluationlayallm-eval
  • Learn decision models by building with them. TypeSafe Jev and Microsoft Decision-1 route support tickets in under a second for about $0.02 per 1,000 calls; Claude handles the hard cases. Interactive site, benchmark, calibration tests and docs.

    shashikantdev3/decision-models-lab · TypeScript

    50
    ai-agentsanthropicbenchmarkcalibrationclaudedecision-modelhuman-in-the-loop
  • Evaluation tools for Jev-compatible decision models, covering accuracy, calibration, hidden instructions, padding and consistency.

    mertkayacs/jevoss · Python

    51
    benchmarkcalibrationdecision-modelevaluationllm-evaluationprompt-injectionpython
  • MCP server and agent skill for the TypeSafe AI System One API (Jev): decompose a judgment into Choice / Score / Noul questions, lint them, measure on labelled data, and put calibrated thresholds in code

    simota/tenbin · TypeScript

    42
    agent-skillscalibrationclassificationclaude-codeconfidencellmmcp
  • Code for 'When Multiple Choices Are Correct: Asymmetric Probability Allocation in Jev and Other Decision Models' (arXiv:2610.29217)

    Rushab1/gr-jev · Python

    30
    arxivcalibrationchaosnlidecision-modelecthrhotpotqallm-evaluation
  • Free English RAG benchmark for TypeSafe Jev 1.13 — frozen candidate pools, calibration, paired bootstrap CIs, $0 runs. Compared with OpenJev and NVIDIA cross-encoders.

    emretheus/jev-rag-benchmark · Python

    30
    benchmarkcalibrationinformation-retrievalllm-evaluationnvidiaopenjevrag
  • Open-source multi-device Jev-like decision models you can train and run yourself: 5.33 ms on one CPU thread. Apache-2.0.

    tamkimd/tamev · Python

    30
    calibrationcoremlcpu-inferencedecision-enginedecision-modeledge-aijev-alternative
  • Measure when to use Jev and other models on your data, then route accordingly.

    FirasSX914/calibre · Python

    20
    pythonbenchmarkcalibrationconfidence
  • Open, CPU-sized Jev-style (System One) decision model: situation + question + options → calibrated probabilities. 122M, ~20 ms on one CPU core, runs in your browser.

    amyrmahdy/decima · Python

    20
    calibrationdecision-modeljev-alternativemultilingualonnxtyped-decisions
  • Decision circuits: typed questions to a System One model, calibrated probabilities back, gates in code. Zero-dependency Python SDK with LangChain, OpenAI Agents, and Claude Agent SDK integrations.

    Barneyjm/decision-circuits · Python

    21
    ai-agentscalibrationdecision-circuitsguardrailslangchainllmpython
  • Runnable samples for Jev, TypeSafe AI's System One model. Send state and questions carrying their own answer options, then branch on the typed value that comes back.

    aarora79/jev-samples

    21
    calibrationclassificationcode-samplesconfidence-thresholdsdecision-makingexamplesllm
  • Jev vs GPT-4.1 as synthetic survey respondents on Twin-2K-500. How you ask mattered more than which model you used.

    jjd-lab/jev-synthetic-survey · Python

    20
    behavioral-economicsbenchmarkcalibrationdigital-twindigital-twinsllmllm-evaluation
  • PyTorch implementation for training JEV-style calibrated decision models

    smha1012/jev-torch · Python

    20
    calibrationdecision-modelhuggingfacemachine-learningpytorchrunpodpython
  • A Claude Code hook that asks whether the decision you are writing needs a model at all. Includes a measured 149-row comparison of TypeSafe Jev against Claude Haiku 4.5.

    wotai-dev/typesafe-jev-tools · TypeScript

    20
    ai-agentsanthropiccalibrationclaude-codeclaude-code-hooksllm-classification

1–50 of 81