# Everlyn-Labs/Everlyn-1

The first open autoregressive foundational video AI model.

Repository: https://github.com/Everlyn-Labs/Everlyn-1
Canonical: https://ross.abutalabs.com/products/everlyn-1
License Family: other
Last push: 2024-10-14T08:09:10+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 49
- inputs: {"age_days": 689, "days_push": 688, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2892, forks 486 (observed 2026-08-28T04:07:28.483914+00:00)

## What it is
Everlyn-1 is an open autoregressive foundational video AI model from Everlyn Labs, accompanied by research on video compression/tokenization (Wasserstein-based vector quantization), efficient autoregressive image/video generation (EfficientARV), and hallucination reduction for multimodal LLMs (ANTRP/TAME). It targets generative video tasks such as animation, inpainting, outpainting, prediction, and interpolation.

## Use cases
- generate videos from text or images with an open autoregressive model
- animate a still image into a video
- inpaint or outpaint missing regions in video frames
- predict future video frames or interpolate between frames
- tokenize and compress video for autoregressive generation
- reduce hallucinated objects in multimodal LLM outputs
- research efficient autoregressive video generation architectures

## When to choose
- you need an open foundational video generation model you can study or fine-tune
- you are researching vector quantization or video tokenization for autoregressive models
- you want to experiment with joint image and video generation tasks
- you are working on hallucination mitigation in multimodal LLMs

## When to avoid
- you need a polished production video-generation service with a stable API
- you require a permissive license for commercial use - no license is specified
- you lack GPU resources for training or running large generative models
- you need a turnkey text-to-video tool rather than research code

## Facets
- artifact type: library
- maturity: experimental
- function: machine-learning, deep-learning, llm-training, video-processing, image-processing, nlp
- domain: artificial-intelligence, deep-learning, large-language-models, image-processing
- platform: python
- tags: video-generation, autoregressive-model, vector-quantization, video-tokenization, multimodal, hallucination-reduction, research-model, video-prediction, video-interpolation, inpainting, wasserstein-vq, efficientarv, antrp, tame, video, natural-language-processing, gpu, linux

## Member repositories
- Everlyn-Labs/Everlyn-1 (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:28.483914+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:48:15.941673+00:00, confidence not recorded.
  - readme: https://github.com/Everlyn-Labs/Everlyn-1 (fetched 2026-08-28T04:07:28.483914+00:00, sha 2ebe6c8fde6e)
- Data as of 2026-08-30T08:39:29.467469+00:00.