First person through smart glasses, as a style LoRA for MiniMax H3.

A wide lens that doesn't bend the world, a head that sways and looks down at what the hands are doing, the wearer's own hands coming in from the bottom of the frame.

How to use

Trigger: povsint

Weight 0.6. At 0.8 the look gets stronger and the hands stop following your description. At 1.0 it overshoots.

Put the style in the first sentence of the H3 prompt: [Shot 1] Live-action in the povsint style, an eye-level shot...

Don't describe the capture (glasses, POV, first person, fisheye). The trigger carries it.

Describe the hands in full, every time: skin tone, rings, watch, nails, sleeves, and "no tattoos" if you want none. Whatever you leave out, the LoRA fills in. Check the hands before you post: they can come out a shade lighter, and now and then a small tattoo slips onto a wrist.

Best in everyday life from behind the eyes: cooking, riding a bike, pulling a book off a shelf, opening the door to a dog. Run it alone, without other anatomy or action LoRAs.

Tested on MiniMax H3 FL2VA pruned int8 ConvRot in ComfyUI, text-to-video: 20 steps, res_multistep, simple, 768x1024, 5 seconds. Trained on FL2VA pruned.

Original style. Use what you make with it however you like, commercially included, and credit the style. Reuploading this LoRA, or training another model on its outputs, needs our permission first: send a message.

Commissions are open: bring me your look or your camera and I'll cut an H3 LoRA to measure. Send me a message.