Every prompt on Faymas works the same way. You take one of your own photos, paste the prompt text into ChatGPT or Gemini, and the AI rebuilds your photo in a new style, outfit, pose, or place. Your face stays yours. The scene around you changes.
If you have never done it before, the whole thing takes about two minutes. Here is the exact process.
Step 1: Pick a prompt you like
Browse the categories and open any prompt. Look at the example photo first. That picture shows the style you will get: the light, the clothes, the mood. If you like the example, the prompt will work with your photo too.
Tap the copy button and copy the full prompt text. Copy all of it, including the last part. Many prompts end with a "negative prompt" section that tells the AI what to avoid, like blurry skin or changed facial features. That part matters. Leaving it out is the most common reason a first result looks off.
Step 2: Choose the right photo of yourself
The photo you upload decides how good the result will be. Pick one that follows these rules:
- Your face is clear and easy to see, looking at the camera or close to it
- Taken in daylight or bright indoor light, not in a dark room
- No sunglasses, cap shadow, or hair covering half your face
- No beauty filter and no heavy editing
- Only you in the frame, especially for your first tries
A plain selfie taken near a window beats a stylish group photo every time. The AI needs to learn your face from this one picture, so give it a clean look.
Step 3: Paste it in ChatGPT
- Open ChatGPT and start a new chat
- Tap the plus or paperclip button and upload your photo
- Paste the full prompt text under the photo
- Send it and wait for the image
ChatGPT reads the photo and the instructions together. Most prompts on Faymas are written for this flow, so you do not need to change a single word.
Step 4: Or paste it in Gemini
- Open the Gemini app or gemini.google.com
- Tap the plus button and upload your photo
- Paste the full prompt text
- Send and wait for the result
Gemini is fast and handles outfit and background changes well. ChatGPT often follows long, detailed instructions a little more closely. Both are good. If one gives you a weak result, try the same prompt and photo in the other before you give up on the prompt.
Step 5: Check the result like a photographer
Look at three things:
- The face. Does it still look like you? Same nose, same eyes, same jaw?
- The hands. Hands are the first place AI slips. Extra fingers or bent fingers show up here.
- The edges. Check where your hair, shoulders, and clothes meet the background.
If the face is right and the style is right, download the photo at full size. If something small is wrong, do not start over. Reply in the same chat with a short fix, like "keep everything the same, fix the right hand" or "same photo, softer light on the face." Small corrections inside the same chat work far better than a fresh start.
If the first result is bad
Regenerate once. AI photo tools vary a lot between tries, and a second attempt often fixes the problem on its own.
Still bad? Change the photo, not the prompt. Nine times out of ten, a weak result comes from a dark, filtered, or group photo. Swap in a clean selfie and run the same prompt again.
If the face keeps changing, add this line at the top of the prompt before you paste it: "Preserve his exact facial identity from the uploaded photo. 100% face match, do not change the face shape, eyes, nose, or lips." Use "her" instead of "his" where it fits. Most Faymas prompts already carry a line like this, which is why they hold faces better than random prompts you find on social media.
A quick word on privacy
Faymas never receives your photo. The prompts are just text. Your picture goes only into your own ChatGPT or Gemini account, under that app's own privacy rules, the same as any photo you share there. Nothing is uploaded to this site when you copy a prompt.
Your first three tries
Start simple. Pick a portrait prompt with a plain background, run it with a clean selfie, and learn how the tool behaves. Then try an outfit change. Then try a full scene, like a travel or bike prompt. By the third one, you will know exactly which photo of yours gives the best results, and every style after that gets easier.
Now pick a category, grab a prompt, and make your first one.