Video
CommunityJev Demos & Experiments
JevGAG Adaptive Temporal Granularity Videos
Hugging Face dataset of 80 generated rollout videos (422 MB) supporting studies on Jev-routed temporal granularity for causal video generation, produced with Causal-rCM on Tesla T4 and A100 GPUs.
- Format
- Video
- Reading time
- 2 min
- Published by
- Community
- Added
- Oct 2026
Highlights
- 80 validation rows totaling 422 MB, licensed CC-BY-4.0 and maintained by MD Jisan Hossen (NUIST).
- Clips generated by Causal-rCM (Wan2.1-T2V-1.3B) at 480x832 and 16 fps on a Tesla T4 and an A100-SXM4-80GB.
- Filenames are prefixed by experiment group, covering online policies, controller ablations, cost/quality sweeps, and hold-mode routing.
- Groups include a100online_*, a100rq3_*, f1_*, f1noisy_*, f4lite_*, holdmode_*, holdfine_*, rq3_*, and rq4_*.
- Companion artifacts include result figures, a T4 online subset, Causal-rCM checkpoints, and code repositories.
Questions about this entry
Short answers drawn from the linked project or page.
It holds 80 validation rows of generated rollout videos, 422 MB in total, produced as qualitative video evidence for adaptive temporal granularity in causal video generation and video world models, where each temporal segment is routed to a fine, coarse, or hold mode.
All clips are produced by Causal-rCM based on Wan2.1-T2V-1.3B at 480x832 resolution and 16 fps, run on a Tesla T4 and an A100-SXM4-80GB.
Prefixes mark experiment groups: a100online_* and a100rq3_* are A100 online-policy and controller-ablation runs, f1_* and f1noisy_* are T4 cost/quality and noisy-context sweeps, f4lite_*, holdmode_*, holdfine_*, rq3_*, and rq4_* are T4 policy, hold-mode, and monotonicity studies.
The dataset is released under CC-BY-4.0, so reuse is permitted with attribution to the maintainer.
It contains 80 rows in a single validation split, with a total file size of 422 MB. The dataset viewer auto-converts it to Parquet and reports 18 downloads in the last month.
Code lives in the GitHub repositories j1s4nn/JevGAG and j1s4nn/jev-adaptive-granularity-WM, with project pages for JevGAG and the Jev-Routed Temporal Granularity paper, plus Zenodo preprints carrying DOIs.
Yes. Result figures are in j1s4nn/jev-adaptive-granularity-figures, the original T4 online subset is in j1s4nn/f4lite-videos, and Causal-rCM checkpoints are in j1s4nn/causal-rcm-ckpts.
More Jev resources like this
A Python demo that screens PubMed abstracts for one systematic review question using Jev. Six narrow questions per abstract go to Jev in a single call, with rule-based tiers and confidence gates deciding include, exclude, or human review.
Jev Demos & Experiments#demo#python#madewithjev
Communityanondo1969/jev-evidence-triage
A static page that races three LLM judging gates — Laya, JEV, and zero-token ZTC — on the same Tetris board with identical pieces, proposals, and noise, publishing paired-match results and the measured constants behind them.
Jev Demos & ExperimentsPlayground#community#playground#demo
Community
A demo that has three decision models read recent news and price Polymarket and Kalshi questions, then compares their probabilities with live market odds and replays how each model changed its mind article by article.
Jev Demos & Experiments#community#github#demo
Community
yorkeccak/model-vs-marketTypeScript
yorkeccak/model-vs-marketTypeScriptThree TypeScript demos built on one pattern for students: Jev answers a typed question with a probability, your code decides what to do, and Claude works only where it is needed. Demos cover Jira triage, pre-flight checks and PR routing.
Jev Demos & Experiments#community#github#demo
Community
soyjorgepilo/jev-demoTypeScript
soyjorgepilo/jev-demoTypeScriptA pygame Flappy Bird that hands the bird to Jev: each new pipe becomes one multiple-choice question about simulated flight paths, and Jev answers with a typed choice plus a probability distribution.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
leftspace89/jevbirdPython
leftspace89/jevbirdPythonA chatbot with no language model: every word is picked from roughly 4,700 options by Jev Choice questions, with confidence-driven lookahead, beam search across sentence directions and a self-critique loop that rewrites weak sentences. Code for the paper and the live demo at talktojev.com.
Jev Demos & Experiments#community#chatbot#research
Community
xucian/talktojevPython
xucian/talktojevPythonFrom 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.