CommunityJev Demos & Experiments
Laya Demo
The official Laya playground: eight tabs that show choice, score, and noul decisions on ticket triage, phishing, guardrails, RAG filtering, moderation, routing, and multilingual text.
- Format
- Playground
- Published by
- Community
- Author
- Convai Innovations
- Last verified
- Oct 2026
- Platforms
- Browser
- Capabilities
- Open sourceRuns locally
Highlights
- Runs on Hugging Face ZeroGPU, so visitors need no local GPU and no account.
- The multilingual tab shows Router picking a checkpoint by script; the card notes Khmer scores 0.000 accuracy at 0.952 confidence on the English model.
- MASSIVE 51-language intent macro-averages: English checkpoint 0.227, multilingual 0.366, against a 0.050 random baseline.
- The email tab is generalization, not a trained skill — the card says the model saw no email data.
Watch out
A preview checkpoint running on shared ZeroGPU quota. The card says it is stronger on routing, classification, moderation, and guardrails, and weaker on rubric scores.
Questions about this entry
Short answers drawn from the linked project or page.
No login and no local GPU: the space runs on ZeroGPU with the usual per-visitor quota. The model files are pulled at runtime, so the space itself stores no weights.
The Router sends non-Latin scripts to laya-multilingual automatically before the forward pass. The card's own example is Khmer, where the English checkpoint scores 0.000 accuracy while reporting 0.952 confidence, so confidence gating cannot catch a script the encoder cannot read.
Eight tabs cover choice, score and noul decisions on ticket triage, phishing, guardrails, RAG filtering, moderation, routing, multilingual text and email, so you can compare behavior across the family's intended tasks.
No. The card says the email tab is generalization rather than a trained skill, because the model saw no email data during training.
On MASSIVE's 51-language intent macro-average the multilingual checkpoint reaches 0.366 against the English checkpoint's 0.227 and a 0.050 random baseline.
Rubric scores. The card says the checkpoint is stronger on routing, classification, moderation and guardrails, and weaker on rubric scores.
No. It is a preview checkpoint running on shared ZeroGPU quota, which is why the card frames it as a preview rather than the released build.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsHacker News: Pointer to a DiffusionGemma plus vLLM tutorial for a Jev-shaped classifier.
1 pointHow to build a Jev-style classifier with DiffusionGemma and vLLMRead the thread on Hacker NewsHacker News: A one-token logprob wrapper that also classifies webcam frames.
allanrbo151 pointsA single function Jev-like wrapper for LLMs, including vision modelsRead the thread on Hacker NewsMore Jev resources like this
A ZeroGPU demo of Julia 1: define a state and any set of choice, score, and noul questions and see per-option probabilities rendered as bars and raw JSON.
Jev Demos & ExperimentsPlayground#community#open-models#playground
Community
221GitHub stars
An agent skill that reproduces Jev-style single-step logits over indexed browser controls using local Qwen weights for offline browser use.
Jev Demos & Experiments#community#browser#open-models
Community
APUS-AI-Lab/fast-browser-usePython
APUS-AI-Lab/fast-browser-usePython306GitHub stars
Local macOS computer-use automation on Apple silicon using on-device laya typed decisions and CUA-S1-FORMS for filling, via the Accessibility API.
Jev Demos & Experiments#community#browser#local
Community
FluidInference/FluidUseSwift
FluidInference/FluidUseSwiftThe Hugging Face hub for Laya's open System 1 checkpoints: the 421M English model at the repo root, a 322M multilingual build in a subfolder, and the typed-decisions fine-tune, all Apache-2.0.
Jev Models & Reimplementations#community#open-models#local
Community
The fine-tuned Laya checkpoint for four typed-decision workflows: 0.766 accuracy on 2,000 held-out decisions, above Jev 1.13.0's published 0.727 and the teacher self-agreement ceiling it trained past.
Jev Models & Reimplementations#community#open-models#local
Community
A browser WebGPU export of Julia 1: the same 551 MB decision checkpoint, a Rust WASM tokenizer, and a JavaScript API that answers typed questions client-side with no server or key.
Jev Models & Reimplementations#community#open-models#local
Community
From the Jev guides
Original write-ups on what this entry is for.
Hundreds of repositories carry the system-one tag, but the ecosystem is a dozen ideas in different sizes. A tour of the families, the licenses, and what a download does not include.
Jev Models & Reimplementations#open-models#hugging-face#local
Read article
Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.
Jev Tools & CLIs#open-models#local#onnx
Read article
Interface, size, license, claims — in that order. A selection guide across the system-one catalog, from a 706K form scorer to 27B Jev students.
Jev Practices & Patterns#open-models#evaluation#benchmarks
Read article
From the Jev community
Posts from builders shipping with Jev right now.