Taken by Phone is an experimental LoRA that adds a touch of everyday realism — the kind that’s not quite polished, just like what you’d get from a budget Android phone camera.

The dataset is made entirely from real photos taken with my own phone, so the results might lean a bit toward Southeast Asian facial features or locations. 😅 That said, you can easily steer the look by adding regional context in your prompt — try keywords like Russian woman, American street, European building, and so on, to match your target vibe.

This LoRA doesn’t require a trigger word, but adding phrases like "taken by phone", "taken by camera phone", "phone camera", or "camera phone" in your prompt can help reinforce the look.

Recommended weight: 0.8 – 1.2

Still a work in progress, so feel free to share your feedback or suggestions!

------------------------------------------

for tensor.art statement:

"This model's training materials do not include any celebrity or unauthorized content. Please do not use this model to post any content resembling celebrities."

------------------------------------------