fal-portrait-lora-skills

Generates and evaluates images from a trained portrait LoRA using fal-ai/flux-lora, including LoRA scale sweeps, local image downloads, and review contact sheets. Use when the user asks to run inference, generate, render, sample, or test images from a LoRA; to check likeness or whether a LoRA is overtrained; or mentions flux-lora, lora_scale, safetensors weights, a trigger phrase in a prompt, guidance scale, image_size, or seeds.

Prepares portrait LoRA training datasets: face-centered 1024x1024 crops, mandatory visual QA and manual crop correction, contact sheets, ZIP packing, and descriptive [trigger] captions via Florence-2. Use when preparing images for FLUX portrait training, fal-ai/flux-lora-portrait-trainer, Kohya, or person LoRA training; or when the user mentions dataset prep, captioning training images, cropping faces for LoRA, or building training ZIP archives.

Validates, uploads, submits, monitors, and persists fal portrait LoRA training runs using fal-ai/flux-lora-portrait-trainer. Use when the user asks to train, fine-tune, start, submit, or monitor a portrait/person LoRA from a prepared ZIP; or mentions images_data_url, trigger_phrase, portrait trainer, training steps, learning rate, LoRA weights, or a fal training request.