Free workflow

Flow Image — stills from a video model

A photoreal still out of MiniMax H3.

H3 has no image mode; this renders the shortest valid clip and keeps its settled middle frame, with a photoreal LoRA. Includes the Advanced variant.

Drag the .json onto ComfyUI. Runs best on the Gmanski RunPod template.

You get

You need

Model packs · Text to image

minimax-h3-image

On the template: gmanski-packs install minimax-h3-image

Pairs well with.

A new photo of a person from a single reference picture. No training.

Clean Krea 2 text-to-image with the sampler that actually works.

The community realism fine-tune, without the confetti and speckles.