User repository on GitHub
- Forks
- 23
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Repos & SDKs
MedJev — JunMa11
Clinical variable extraction from free-text notes that fits on one consumer GPU inside your hospital: answers thousands of notes an hour with typed noul, choice and score outputs, a confidence per answer, and no per-use fees.
- Stars
- 114
- Language
- Python
- License
- APACHE-2.0
- Reading time
- 2 min
- Published by
- Community
- Author
- JunMa11
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open sourceRuns locally
Highlights
- Three task types: noul (yes/no returned as a probability), choice (one of 2-255 named options), and score (an ordered level).
- The README claims thousands of notes an hour on one consumer GPU, with no subscription or per-use fees.
- Patient text never leaves your network; the README notes an optional hosted Jev baseline from TypeSafe AI.
- Every answer carries a confidence, so clear-cut charts can be accepted automatically and doubtful ones sent for human review.
- Adding a new field needs a few hundred labelled example charts and a few hours of retraining, per the comparison table.
Quickstart
bash
uv venv --python 3.12
source .venv/bin/activate
pip install torch==2.14.0 --index-url https://download.pytorch.org/whl/cu130
pip install -e .Watch out
README reviewed 2026-10-17; the excerpt is truncated inside the optional TypeSafe Jev baseline section and lists no benchmarks, deployed results or licence text beyond the Apache-2.0 metadata, so the throughput and accuracy claims are the project's own.
Questions about this entry
Short answers drawn from the linked project or page.
It turns free-text clinical notes into table-ready data by answering predefined typed questions: yes/no returned as a probability (noul), one of 2-255 named options (choice), and ordered levels (score), with a confidence attached to every answer.
Its README offers an optional hosted Jev baseline from TypeSafe AI, and its noul, choice and score answer types follow the same typed-decision pattern: every field is filled with one of the answers you defined rather than free-form generated text.
Create a Python 3.12 environment with uv venv --python 3.12, activate it, install torch 2.14.0 from the cu130 index, then run pip install -e . in the repository. The README also shows a check command that verifies torch, transformers, peft and fla import and whether CUDA is available.
The README says it is small enough to run on one consumer GPU inside your own hospital and claims thousands of notes an hour. The install instructions target a CUDA build of torch and the verification step runs nvcc, so a CUDA-capable GPU is expected.
The README says no subscription or per-use fees, because everything runs on your own hardware. The only hosted component mentioned is an optional Jev baseline from TypeSafe AI, which the truncated excerpt does not price.
Its comparison table says it only answers the fields it was built for: it does not summarise, explain, draft text or answer open-ended questions, and adding a new field requires a few hundred labelled charts plus retraining.
Apache-2.0 per GitHub metadata, written in Python, created and last pushed 2026-09-22 with 114 stars and 23 forks. The digest shows no releases, changes or benchmarks beyond the README.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.
FacebookClinicnotes
Other listings of this project
Forks and copies of this project in the directory. This entry is the canonical one.
MEDJEV is an independent, JEV-inspired research implementation for typed evidence decisions: a clinical record or literature passage, a runtime question and a candidate set supplied at call time, scored into an explicit probability distribution plus calibration scope. Alpha research code, not clinical software.
Jev Practices & Patterns#community#awesomejev#catalog-import
Community
PAI-CUHK/MEDJEVPython
PAI-CUHK/MEDJEVPythonMore Jev resources like this
A System One layout optimizer for decision-model workloads: it reorders rows and fields so a prefix-caching backend reuses input computation, then returns per-record decisions in original order. Its bundled backend serves AutoTrust JEV-9B through vLLM.
Jev Repos & SDKs#community#github#python
Community
0814wdwd/solo_jevPython
0814wdwd/solo_jevPythonA Bangla-focused adaptation of Laya, the non-autoregressive typed-decision model: targeted data preparation, fine-tuning and evaluation for classification, routing and scoring on Bangla, Banglish and code-switched text, with the checkpoint still to come.
Jev Repos & SDKs#community#github#python
Community
shamspias/laya-banglaPython
shamspias/laya-banglaPythonHTTP server for the Mica v0.5 4B Preview decision model: GGUF weights run through llama.cpp and answer noul, choice and score questions with calibrated probabilities over the TypeSafe /v1/systemone format.
Jev Repos & SDKs#community#github#python
Community
akivet/MicaPython
akivet/MicaPythonServing code for the Jeff 1.0 family of Jev-style decision models: a state, a typed question and its options go in, and one forward pass returns a probability for each option. Four open models from 4B to 31B parameters are served the same way.
Jev Repos & SDKs#community#github#python
Community
j-geuter/jeff-servePython
j-geuter/jeff-servePythonA frozen decision server that serves Gemma 4 31B — quantized to FP8 at startup — in TypeSafe's /v1/systemone format, reading the next-token distribution instead of generating tokens. It needs one FP8 NVIDIA GPU with 80 GB and has no authentication.
Jev Repos & SDKs#community#github#python
Community
krishna-gogineni-765/deck31bPython
krishna-gogineni-765/deck31bPythonAn open-source, locally executed decision engine in the spirit of Jev (TypeSafe AI): state becomes features, and Noul, Choice and Score heads return typed, calibrated probabilities without generating text. It ships a cheap-to-strong cascade and a full train-calibrate-evaluate-deploy cycle.
Jev Repos & SDKs#community#github#python
Community
slavabars/decision-cascadePython
slavabars/decision-cascadePythonFrom 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.