Files
hao-ai-lab-FastVideo/scripts
Aryan Kumar a224c577b0 [feat]: Pika H3 Flex Ref2VA coverage rows from the OmniRef v4 sources
Re-encode references and Qwen3-VL conditioning with Flex's tier sizing (1024 short edge
below 768p) and add audio-reference, mixed image+video, 6-9 image and two-video rows,
reusing the v4 target latents.
2026-10-29 22:13:55 -07:00
..