Best Jev Projects

Search 530 verified repositories across 11 categories and 25 languages.

Last updated
Submit project
Structured inputs flow through Jev into probabilistic decisions

ProjectLanguageAddedDescription

  1. different-ai/openwork

    Agent tooling    23,654TypeScript

    Open-source Claude Cowork alternative whose eval testkit can use Jev as a typed verification judge for agent-produced work.

  2. browser-use/jev-ultrafast

    Browser & computer use    9,291Python

    Browser Use's ultrafast agent: Jev picks the operation and DOM element in one request; a small LLM only writes text when typing is needed.

  3. vercel/eve

    Frameworks & integrations    5,274TypeScript

    Vercel's open agent framework, which ships Jev as the default evaluation model in its experimental evaluate path.

  4. tamaratran/fast-jev-compaction

    Agent tooling    4,427TypeScript

    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 dropped or truncated, everything kept stays verbatim.

  5. TheoLeeCJ/SemIf

    Jev-like models    2,023Python

    Semantic ifs from open models, on a 3090 at home. Independent; not affiliated with Jev or TypeSafe.

  6. jaredpalmer/kev

    Jev-like models    1,558Python

    Trainable family of small Qwen-based Jev-like decision models with typed primitives, datasets, evaluation tools, and local inference.

  7. jarrodwatts/jev-trader

    Applications    1,349TypeScript

    One AI trade decision every Monad block. Jev on Kuru MON-USDC.

  8. TianyuCodings/NanoJev

    Benchmarks & research    1,074Python

    Open 0.6B Jev replica with parallel decisions, complete probability distributions, training pipeline, weights, dataset, and live demos.

  9. vinnylarouge/jevlike

    Benchmarks & research    1,008Python

    Train a small model that chooses among a changing list of text options, one probability per option in a single pass. Includes Doom, chess, and Wikispeedia demos.

  10. vercel-labs/ai-cli

    Frameworks & integrations    807TypeScript

    Vercel Labs terminal CLI that can run Jev as the evaluation model for its evaluate command.

  11. typesafe-ai/skills

    Official    765—

    Official agent skill for Claude Code, Codex, and compatible agents: primitives, patterns, and how to structure evaluations.

  12. githubnext/localjev

    Jev-like models    662TypeScript

    Local Jev-compatible System One API that converts typed questions into DiffusionGemma classification prompts and probabilities.

  13. Anil-matcha/awesome-jev-by-typesafe

    Other lists    641Python

    Evidence-backed Jev use cases, patterns, prompts, and starter code.

  14. awlevin/typesafe-computer-use

    Browser & computer use    537Python

    Computer use for about $0.0002 a step: OCR the screen, classify the next action with TypeSafe, click. macOS.

  15. featherless-ai/simple-jev

    Jev-like models    429Python

    Open-model Jev-style server that reads next-token logits to return typed choices, rubric scores, and truth judgments.

  16. smithersai/smithers

    Frameworks & integrations    417TypeScript

    Agentic TypeScript workflow framework with a Jev session checker wired into its workflows.

About the catalog

What is bestjev?

bestjev is an independent, community-maintained directory of open-source projects built around Jev, TypeSafe AI's System One model for typed decisions.

Each listing links to its original repository and records its category, primary language, GitHub star snapshot, and a localized description, helping developers find verifiable Jev SDKs, agents, integrations, applications, research, and tools.

bestjev is not affiliated with or endorsed by TypeSafe AI.

What is Jev?

Jev is TypeSafe AI's System One model for decisions that software needs to consume. Instead of generating free-form prose, it accepts a state and one or more typed questions, then returns structured answers with calibrated probabilities.

It fits workflows where the option space or scoring rubric is known in advance: routing an agent, validating a tool call, ranking candidates, checking a policy, or deciding whether an action should proceed.

Operational advantage

Applications can set deterministic thresholds, inspect uncertainty, and route low-confidence cases to a person or a larger model.

CHOICE

Pick one option

Select from a known set and return probabilities with confidence.

SCORE

Rate on a rubric

Place an input on an ordered scale with a full distribution.

NOUL

Estimate truth

Return the probability that a statement is true, from zero to one.

How is this catalog verified?

Each entry is checked against public evidence: the upstream repository, project-authored documentation, and visible Jev usage. Stars are a dated discovery snapshot, not a quality ranking.

Curated by

“Think of Jev as a frontier-intelligence function call: unstructured state in, typed probabilistic decisions out.”— TypeSafe AI, introducing Jev

Which sources does bestjev use?

Contribute

How to submit a new project

Know a public project that uses Jev? Send us its repository and the evidence needed to verify it. Submissions are reviewed before they appear in the catalog.

Submit a projectOpens a pre-filled GitHub issue.
  1. 01

    Share the repository

    Provide the public GitHub URL and a concise description of what the project does.

  2. 02

    Show how it uses Jev

    Link to code, documentation, a demo, or another public source that makes the Jev integration verifiable.

  3. 03

    We review and publish

    We verify the evidence, choose the right category, and add accepted projects in a future catalog update.