Show defining features clearly.
Use a sharp, unobstructed face image with natural proportions. Heavy blur, masks, extreme angles, or inconsistent ages make identity guidance less reliable.
Create a reusable character from face and full-body references, then select that character again when you generate an image, recreate a scene, direct a storyboard, or make a reference-led video. TongAI keeps the references connected to one saved profile so you do not need to rebuild the setup for every scene.
Create a reusable character Saved characters · Face and body references · Image and video workflows
TongAI has no prompt restrictions, including for adult creative themes. Choose the subject, tone, and visual style that fit your idea, and write the direction in your own words.
One identity, reusable inputs
A TongAI character is a saved reference profile, not a claim that every output will be pixel-identical. Add clear images that show the face and body, give the character a recognizable name, and select that same profile in compatible workflows. The selected model receives the references as guidance while your prompt supplies the scene, action, camera, wardrobe, and visual style.
This is useful when you want the same person or designed subject to appear across portraits, environments, storyboard frames, or short video takes. It also makes iteration simpler: keep the identity source stable, change one creative variable at a time, and compare each result before continuing.
Use a sharp, unobstructed face image with natural proportions. Heavy blur, masks, extreme angles, or inconsistent ages make identity guidance less reliable.
Add a standing or otherwise readable full-body view when body shape, height, outfit, or physical design should remain recognizable beyond close-ups.
Avoid combining conflicting faces, hairstyles, wardrobes, or art styles. Reuse the same set and repeat essential identity details in the prompt.
Connected workflows
The compatible model and route determine which references are accepted. TongAI shows those inputs and the exact credit price before generation.
Upload clear face and full-body references, add a name, and keep the reference set available for future image and video work.
Create a characterUse a saved character in Recreate an Image, add a target scene or prompt, and guide the model with the same identity references.
Recreate an imageChoose a compatible character-reference video model and direct the action, camera, setting, wardrobe, and sound in your own words.
Create character videoIn Video Director, carry the same saved character into storyboard scenes and review each opening and closing frame before generating a shot.
Explore Video DirectorEvery generation is a new model interpretation. A familiar face can shift with a strong angle change, distant framing, fast motion, occlusion, unusual light, or a prompt that contradicts the reference. Video adds another challenge because the identity must remain coherent across many frames.
Paid generation without a fixed daily cap
TongAI does not impose a fixed daily generation limit. Image, scene, and video requests use credits at the rate shown for the selected model and settings. Video is commonly priced per generated second; images and recreated scenes are priced per completed output. Keep creating while you have enough credits.
See current image and video pricingSave a character with a clear face and, when useful, full-body references. Select that same character in a compatible image workflow, repeat important identity details, and keep the requested framing and lighting readable. Each result still needs review.
Yes. A saved TongAI character can be selected in compatible image, scene, storyboard, and character-reference video workflows. The exact inputs and consistency behavior depend on the chosen model.
Yes—paid generation has no fixed daily cap while you have enough credits. It is usage-based: every request has a displayed price, and you can add credits whenever you need more.
No. Face, body, image, video, and LoRA references guide compatible models, but they do not guarantee exact identity, anatomy, clothing, motion, or composition.
Save the identity references, then use them wherever the selected model and workflow support character guidance.