Pika H3 Flex PTD

Model Nov 1, 2026

WAS's community model patches combine Pika H3 Flex adapters with four-step MiniMax H3 PTD files for a ComfyUI workflow producing video and audio.

Visit primary link 2 sources attached

Frame from WAS's generated lighthouse video example, showing waves, flying birds and a lighthouse at sunset. — Frame at two seconds from WAS's author-published Pika H3 Flex PTD first-frame-to-video sample. This is generated video imagery, not a real location photograph or an AITrending reproduction; it does not measure generation speed. Source: WAS / Pika H3 Flex PTD

self-hosted multimodal

What happened lately

Pika H3 Flex PTD timeline

Ecosystem Nov 1, 2026

Pika Flex gets a four-step ComfyUI video workflow with sound

Community reports 2 sources

The first patch weights were uploaded October 29. On November 1, WAS added an example workflow, input pictures and rendered clips. The files combine Pika H3 Flex adapters at strength 0.5 with four-step PTD files, using separate patches for MiniMax H3 first-frame-to-video and reference-to-video transformers.

The English card documents installation in ComfyUI’s ptd or model_patches directory and loading through H3 PTD Patch in WAS Node Suite. Matching MiniMax H3 transformers, text encoder and video/audio VAEs remain necessary. The files use the MiniMax H3 Community License Agreement. Their four-step behavior and examples are author-documented, not an AITrending performance test.