User repository on GitHub
Run Jev-style typed decisions locally on your Mac with low RAM usage and fast responses
- Forks
- 3
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Demos & Experiments
Laya Mps
Local server running Jev-style typed decisions on a Mac GPU through Metal, with a demo page, a latency benchmark, and tunable memory modes around one FP32 model.
- Stars
- 22
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- afshinm
- Last verified
- Oct 2026
About this project
Run Jev-style typed decisions locally on your Mac with low RAM usage and fast responses. Laya delivers typed decisions with 32 ms median latency using 2.1 GiB RAM on M5 Pro, with a slower 0.74 GiB mode for lower memory use. MPS stands for Metal Performance Shaders, which PyTorch uses to run Laya on your Mac's GPU.
Quoted from the project's README. Read the full README.
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- Python
Highlights
- Serves choice, score, and true-probability questions over a local HTTP API with no key.
- Default mode answers at tens of milliseconds with around 2 GiB process RAM on the test machine.
- Low-memory mode drops resident RAM substantially by streaming layers and embeddings from disk.
- Demo and benchmark pages plus metrics and diagnostics flags expose timing and memory per call.
Watch out
Figures are tied to fixed Pong inputs on the measured machine, and multiple questions can need multiple forwards with extra disk reads. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
The default reduced mode holds transformer layers in RAM with embeddings on disk; minimal streams layers too for lower RAM at higher latency, and full keeps everything resident.
Yes for decisions: all three settings agreed on every one of the 260 benchmark decisions including probabilities, differing only in RAM and latency.
Apple Silicon with a recent macOS, plus disk room for the runtime and model download; later runs reuse local files and inference works offline.
No. It specializes in customer service, invoices, security incidents, and agent traces, and is explicitly not a general-purpose language model.
An Apple Silicon Mac (M1 or newer) running macOS 14 or later, Git and uv, and about 4 GB of free disk space for the runtime, download cache, and the roughly 843 MB model. uv installs Python 3.12 if needed.
A local server on 127.0.0.1:8000 with no API key: POST /v1/decisions for typed answers, plus GET /health, /v1/config, /docs, /openapi.json, and /demo/. Each request accepts 1-16 named questions.
The default model allows 1,024 formatted tokens per question and the optional English checkpoint allows 512; inputs that would be shortened are rejected. Choice takes 2-26 labels, score 2-10 levels, and noul an optional false/true criteria.
More Jev resources like this
46GitHub stars
Open 150M encoder that reproduces the typed noul, choice, and score interface in one forward pass, trainable on a free GPU and runnable locally for nothing.
Jev Demos & Experiments#community#jevbest#catalog-import
941GitHub stars
Library that adds schema-guaranteed typed outputs to existing autoregressive LLMs without changing weights, served through SGLang or a hosted API.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
TypeLLM/TypeLLMPython
TypeLLM/TypeLLMPython45GitHub stars
Drop-in service that exposes Upstage Solar models through the System One shape, accepting state plus noul, choice, or score questions.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
hunkim/solar-mini4-jevPython
hunkim/solar-mini4-jevPython38GitHub stars
Research recipe that fuses two frozen open readers into Jev-style yes/no judgments with a fit-free router, confidence gate, and candidate sets.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
FeiLiuEM/open-medical-jevPython
FeiLiuEM/open-medical-jevPython62GitHub stars
Simulation study that drives one xArm7 apple-to-plate task with Jev, a large reasoning model, and a small model through identical incremental controls.
Jev Demos & Experiments#community#jevbest#catalog-import
40GitHub stars
Open research toolkit that encodes an image once and answers multiple typed choice, noul, and score questions from a shared distribution, plus a local 0.6B text-decision model.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
IamBusy/OpenJev-VisionPython
IamBusy/OpenJev-VisionPythonFrom the Jev guides
Original write-ups on what this entry is for.
A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.
Jev Demos & Experiments#gaming#agents#control
Read article
From the Jev community
Posts from builders shipping with Jev right now.