Versions

A concise history of the MediaAI pipeline and its browser experiences.

v0.8

LTX-first collections

Curated Top 100 lists can now be planned and generated as bounded, GPU-safe LTX batches.

  • Top 100 Cars launch catalogue
  • Exact ready, complete, and blocked coverage before generation
  • Five-button TikTok-style footer across every public MediaAI page
  • All generation remains stopped until an explicit scoped run
v0.7

Complete public library release

Every validated local MediaAI asset is published through one shared S3 and Mongo library for MediaAI Scroller, standalone Scroller, and Wikai.

  • Full completed-asset snapshot instead of the historic 500-per-kind cap
  • Shared playable image, audio, motion-image, archive, and LTX URLs
  • Canonical library, article sidebars, ranker, travel WBP, and scale roadmap
  • Generation remains independently stopped until an approved measured pilot
v0.6

Frame TV release

Production Scroller and VJ become playable directly from the hosted MediaAI site.

  • Random, latest, and ranked article Scroller modes
  • Desktop landscape media contained inside the portrait Scroller frame
  • VJ previous, next, random, keyboard, fullscreen, and expandable article controls
  • Audio library and WIKAI-first article links
  • Read-only pages with no automatic refresh
v0.5

Observable media factory

Local dashboard, live gallery, article pages, Scroller prototype, worker status, and safe restart controls.

  • Independent motion-image accounting
  • LTX, audio, image, articles, VJ, and 10K live views
  • Hourly status and realistic ETA reporting
v0.4

Distribution pipeline

Completed assets gained S3, MongoDB, GitHub index, WIKAI BO, and snapshot publishing workers.

v0.3

Article grounding

Wikipedia article text, source imagery, local thumbnails, Piper narration, and per-topic metadata.

v0.2

Motion and LTX

Independent motion-video generation and the first LTX image-to-video outputs.

v0.1

Local generation baseline

SQLite-backed, restart-safe SDXL image production on the local GPU.