I made videos with Breaking Bad characters and some of the characters I didn't find. I had to train a couple of my own lora :) here is a link to the video with this lora:

This Lora was trained on 15 images of the character Tuco Salamanca (actor Raymond Cruz)

Prompt: (masterpiece, best quality, awards winner) Tuco Salamanka, breaking bad, 1man, professional photography, high resolution, 8k, detailed photo, hyperealistic, ultrasharp, studio light, soft shasows, <lora:tuco:1>

Negative promt: render, 3d, 2d, drawing, digital art, painting, vector art, paint, draw, sketch, disfigured, deformed, doll, fake, photoshop, collage, old, child, wrinkles

Model: Realistic Invision

Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 3268175131, Size: 512x704, Model hash: a8e08b582e, Model: realisticVisionV30_v30VAE, Lora hashes: "tuco: f987fc5e84da"

You can use other prompt and other models it gives interesting and unpredictable results