Generate images and videos, pick the best models, use AI editing tools, and copy proven prompts to get great results fast.
Sign up / Sign in (30 seconds)
To use PicLumen, you’ll need an account. Signing in helps you save generations, access history, and sync across devices.
How to sign in
- Click Launch PicLumen
- Choose Sign in with Google (fastest) or email
- You’ll land on the Create page automatically
Understand PicLumen pages (so you don’t get lost)
PicLumen has a few key pages. Here’s what each one is for:
- Home: overview of images, videos, models, and tools
- Studio (Create): where you generate images and videos
- Hub🔥: step-by-step guides + copyable prompts
Core settings (models & parameters)
Start with defaults. If you want better results fast, adjust only one thing at a time.
Model (how to choose)
Pick a model based on what you’re making:
- Realistic → portraits, product photos, lifestyle scenes
- Anime / ACG → characters, manga panels, clean lineart
- Design / Illustration → posters, icons, concept art
- Video models → motion-heavy clips or cinematic storytelling
Aspect ratio (quick picks)
Aspect ratio affects composition and where your content will be used.
- 1:1 → social posts, product thumbnails
- 2:3 → full-body characters, Pinterest visuals
- 16:9 → YouTube thumbnails, banners, cinematic scenes
- 9:16 → TikTok / Reels / Shorts
Number of outputs
Generate multiple images when exploring ideas. Generate fewer when you already have a clear target.
- 4 outputs → best for exploration
- 1 output → best for speed and iteration
Advanced settings (optional)
Advanced settings are optional. Keep defaults unless you know what you want.
- Quality / Steps: higher = more detail (slower)
- Guidance / Prompt strength: higher = follows prompt more strictly
- Seed: reuse it to keep composition consistent
- Reference strength (Image → Image): preserves more of the input image
Advanced Parameters
- Negative Prompt: specify what you don’t want to show in the generated image.
- Guidance Scale: Control how strongly your text prompt affects the generated image.
- Steps: Determine the number of steps used to create an image; more steps can lead to more detailed results.
- Seed: The random seed value controls the initial noise of the latent image and, hence, the composition of the final image. Using the same seed with the same parameters will produce the same image, allowing for reproducibility.
Notice:
- Each of our models has well-tuned optimal original advanced parameters that normally do not need to be altered. If the generated results are not sufficiently as expected, the users can adjust advanced parameters to get better results.
- IMPORTANT! If users click remix, then all the generated images will keep the parameters the same as the ones you choose to click remix. If you need the advanced parameters back to the original parameters, click “reset”.
Generate images (2 main ways)
Text → Image
Use Text → Image when you want to create from scratch.
Best prompt structuresubject + style + lighting + camera + background
Image → Image (Reference)
Use Image（Add Photos） → Image when you want variations while keeping composition or style.
Best for
- consistent characters
- style variations
- redesigning an image without losing structure
Content Reference:
It will keep the core content of the uploaded image intact, combining your text prompts to change colors, restyle the image, turn sketches into masterpieces, etc.
For example, on the left is the uploaded image, and I choose “Content Ref,” then I enter the prompt “woodblock print, people standing on the beach, strong contrasting colors, a town on fire.” Then, I get the result on the right.
Style Reference:
PicLumen will mimic the style of the uploaded image, blending it with your text prompts to generate a new stunning AI art.
Character Reference:
This will maintain consistent characters across your images with ease. Use our image to image ai generator to give images a new expression by combining text prompts without changing the characters.
Note: For character reference, users should upload an unobstructed picture of the face.
Generate videos-Hot🔥 (2 main ways)
Text → Video
Describe a scene and generate a short video with motion and cinematic feel.
Image → Video
Bring a still image to life with subtle motion and camera movement.
AI editing tools (polish results in seconds)
Don’t regenerate from scratch. Use tools to fix and refine what you already like.
AI Replace
Replace an object or area with a text instruction.
Best for
- removing unwanted elements
- changing outfits / backgrounds
- fixing small mistakes
Image Extender
CTA: Open Image Extender
Background Remover
Remove background in one click.
Image Upscaler
Increase resolution and sharpness for final export.
Image Colorizer
Restore color to old photos, sketches, and line art.
Remix (iterate faster)
Remix lets you reuse a result as a starting point—so you can iterate faster without rewriting everything.
Best for
- turning a good image into 10 better ones
- exploring styles quickly
- keeping the same idea with different looks
Prompt Enhancement (optional)
Prompt Enhancement rewrites your prompt to make it more detailed and model-friendly.
Use it when
- you’re new to prompting
- your result looks too simple
- you want richer lighting and details
Turn it off when
- you need strict control
- you’re matching an existing character/style
Upload your image
Upload an image to:
- generate variations (Image → Image)
- animate it (Image → Video)
- edit it (Replace / Extend / Remove BG)
Prompt library (copy & paste)
Copy a proven prompt below, generate a result, then tweak one keyword at a time.
Prompt 1 — Product Shot (E-commerce)
Prompt
Minimalist product photo of a premium sneaker on a white seamless background, soft studio lighting, crisp shadow, ultra sharp, commercial photography
Recommended: Realistic / product-friendly image model
Prompt 2 — Cute 3D Character (Full body)
Prompt
A cute 3D cartoon girl standing full body, soft pastel outfit, gentle smile, clean background with subtle bokeh, studio lighting, high quality
Recommended: 3D / stylized model
Prompt 3 — Cinematic Scene
Prompt
A lone traveler walking through a neon rainy street at night, cinematic lighting, volumetric fog, 35mm film look, dramatic composition, ultra-detailed
Recommended: cinematic-friendly model
Prompt 4 — Poster / Graphic Design
Prompt
A modern event poster design, bold typography, clean grid layout, high contrast, minimal graphic shapes, professional print-ready style
Recommended: illustration/design model
Prompt 5 — Text → Video (Food / Lifestyle)
Prompt
A hyper-realistic close-up of a glass of whiskey with ice cubes. A single drop of water is sliding down the side of the cold glass. Dramatic bar lighting, dark mahogany background.
Recommended: FLUX.2 Pro
Prompt 6 — Image → Video (Animate a still)
Prompt
The camera slowly pans around the glass in a circular motion. A hand reaches in and drops a fresh ice cube into the liquid, causing a realistic splash and ripples. 4k, cinematic slow motion.Recommended: Kling 3.0
FAQ & quick fixes
My image looks “plastic”. What should I do?
Use more natural keywords like:soft natural light, realistic texture, subtle skin details
Avoid stacking too many “beauty” words.
The model ignores my prompt.
Keep your prompt short. Put the most important info first.
How do I remove unwanted text or logos?
Add negatives like:no text, no watermark, no logo
How do I keep the same character?
Use Image → Image reference, keep prompt structure consistent, and reuse seed if available.
My video motion looks weird.
Try subtle motion first. Avoid too many actions in one sentence.
The image is blurry.
Use Upscaler, and add:ultra sharp, high detail, crisp edges
