von — security grade SAFE, quality 63/100

Security audit verdict: SAFE · quality 63/100

No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →

by wfzyx · AI Tool · ★ 78

Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h

🔒 Is von safe to install? View the security audit →

About von

🪨 Von () The Open-Source System One Decision Model. Sub-15ms decisions. 14MB footprint. Zero cloud latency. Zero API keys. Zero VC tax. Named in homage to John von Neumann and Ludwig von Mises. Why Von? TypeSafe AI raised $40M to charge $0.042/1M tokens for what amounts to a smart statement behind a closed-source cloud API waitlist. Von is 100% open-source and runs locally on your machine. Powered by the 3.x Simple Attention Network (SAN), Von strips out autoregressive text generation bloat to deliver pure, structured, calibrated decisions in under 15 milliseconds on a standard CPU. Non-Autoregressive: Evaluates all questions in a single forward pass. Zero Hallucinations: Structurally guaranteed output types. No Markdown drift, no JSON formatting errors. Epistemically Calibrated: Confidence scores and probability distributions that reflect statistical reality. 14MB Binary: Runs in 28MB RAM on CPU. No GPU required. Drop-in Jev Compatible: Ships with an in-process SDK and a HTTP server matching TypeSafe's wire protocol.

decision-modeljevmachine-learningpythonrlcdsystem-onetypesafe

Quick Facts

Stars78
Forks12
LanguagePython
CategoryAI Tool
LicenseApache-2.0
Quality Score62.6443967542663/100
Open Issues2
Last Updated2026-09-20
Created2026-09-18
Platformspython
Est. Tokens~15k

von alternative? Top 6 similar tools

Looking for a von alternative? If you're comparing von with other ai tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.

  • openJev-verdict-2.0 by Heman10x-NGU · ⭐ 133

    Calibrated 151M Non-Autoregressive Decision Engine beating TypeSafe Jev & Laya on LocalLLaMA/typed-decisions (

  • awesome-jev-gallery by OmniJev · ⭐ 76

    🔥🔥 Papers, open reproductions and independent evaluations behind System One models and Jev.

  • awesome-jev by OmniJev · ⭐ 65

    🔥🔥 Papers, open reproductions and independent evaluations behind System One models and Jev.

  • awesome-typesafe by AbdelStark · ⭐ 372

    A curated list of official resources and community projects for TypeSafe, System One models, and Jev.

  • awesome-jev-projects by logicrw · ⭐ 177

    Awesome Jev: source-backed open-source ecosystem radar, plain-language project discovery, and automatic GitHub

  • awesome-jev-typesafe by valentynkit · ⭐ 97

    Typed decisions with TypeSafe's Jev, the first System One model

More AI Tool Tools

Explore other popular ai tool tools:

View all AI Tool tools →

Popular Python Agent Tools

Frequently Asked Questions

What is von?

von is The open-source System One decision model. Sub-15ms, non-autoregressive, local drop-in alternative to TypeSafe Jev.. It is categorized as a AI Tool with 78 GitHub stars.

What programming language is von written in?

von is primarily written in Python. It covers topics such as decision-model, jev, machine-learning.

How do I install or use von?

You can find installation instructions and usage details in the von GitHub repository at github.com/wfzyx/von. The project has 78 stars and 12 forks, indicating an active community.

What license does von use?

von is released under the Apache-2.0 license, making it free to use and modify according to the license terms.

What are the best alternatives to von?

The top alternatives to von on Agent Skills Hub include openJev-verdict-2.0, awesome-jev-gallery, awesome-jev. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.

How this security grade is produced

Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.

The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.

Sources & who's responsible:

View on GitHub → Browse AI Tool tools