LaunchVisual AIOctober 30, 2026

Pika releases open-source Pika H3 Flex video model

Read original source →huggingface.co

Pika H3 Flex debuts as the #1 open-weight model on DesignArena's text-to-video board, post-trained on MiniMax H3 to generate in six denoising steps instead of 50. It outputs up to 4K with five editable audio stems and is priced from $0.01/sec.

How this story unfolded

6 days · 1 report · 8 community posts · from Oct 29

  1. Oct 29
  2. Oct 30
  3. Nov 3
  4. Nov 4

More stories today

Open the live feed