Skip to content
JevDirectory.orgTraffic count unavailable. View site traffic

Jev directory

CommunityJev Repos & SDKs

SystemOne Builder

An open Apache-2.0 toolkit to train, benchmark and serve System One decision models on your own hardware, with Kenning as the flagship model. Models answer typed questions with calibrated probabilities behind a POST /v1/systemone endpoint.
See highlightsOpen github.com
Stars
1
Language
Python
License
APACHE-2.0
Reading time
2 min
Published by
Community
Author
systemonedev
Last verified
Oct 2026

systemonedev/systemone-builder

Organization repository on GitHub

View on GitHub

Train, benchmark and serve your own System One decision model: typed questions in, calibrated probabilities out. Runs locally, Apache-2.0.

Forks
0
Last pushed
Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

Tagscommunitygithubpythontraining
Works forJev for ClassificationJev for Evaluation
Used inJev for Software & DevTools
Languages
Python
Platforms
APIBrowser
Capabilities
Open sourceSelf-hosted

Highlights

  • Serves models locally behind POST /v1/systemone; the wire format is compatible with TypeSafe's and never trains on its outputs.
  • Kenning v0.6 is a ~400M long-context ModernBERT cross-encoder scoring 0.761 macro on the project's 1,328-item general suite, vs Jev's 0.830.
  • Train on your own problem from teacher-written cases, labelled data or Clef distillation; 10% is held out for Verify.
  • Quickstart needs one NVIDIA GPU and Docker Compose v2.20+; first up downloads about 5.4 GB.
  • Apache-2.0 code and Kenning weights; the README says it is not affiliated with or endorsed by TypeSafe AI.

Quickstart

bash
cp .env.example .env
docker compose up -d

Watch out

README and GitHub metadata reviewed 2026-10-17; the digest truncates the quickstart. Benchmark tables are the project's own suite and the README itself cautions that its general suite rewarded a change which did not generalise, so numbers are development signals.

Questions about this entry

Short answers drawn from the linked project or page.

From the Jev guides

Original write-ups on what this entry is for.

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai