If you’re opening Artifox for the first time and just want a solid result fast, don’t overthink it:
- Start with AI Tools Hub: upload a photo, generate in one click, get your first win.
- When you want more control (and repeatable outputs), move to Image Creation: start from templates, then tweak prompts.
Real entry points used in this guide:
Pick a workflow in 30 seconds
Answer one question: Do you want speed or control?
- I want results fast (no prompting, upload-first)
- Workflow A: AI Tools Hub
- I want creative control (repeatable, series-ready)
- Workflow B: Image Creation
Workflow A: AI Tools (upload-first, instant results)
AI Tools are the easiest way to start because you don’t need to write prompts or understand parameters.
Pick the tool that matches your goal
A simple way to choose:
- Want a better profile photo or avatar? Start with avatar-focused tools.
- Restoring an old photo? Use restore/enhancement tools.
- Unsure? Pick the tool whose preview examples look closest to your desired output.
Upload a photo: 3 details that matter
Most “bad results come from messy inputs. Try to upload photos that are:
- Sharp (not blurry)
- Normally lit (not crushed shadows / blown highlights)
- Subject-focused (subject takes ~60% of the frame)
Generate & download: get a usable version first
Your first generation goal isn’t perfection—it’s a usable baseline.
- Generate once
- If the direction is right, tweak one thing (more natural, clearer, closer to identity, etc.)
- Download and save
The fastest way to improve outputs:
- Speed first, polish later
- Change one variable per iteration so you know what actually helped
Get your first win with AI Tools
No prompting needed—upload a photo and generate instantly.
Open AI ToolsWorkflow B: Image Creation (templates + prompts, more control)
Use Image Creation when you want:
- A series of consistent images (same vibe/character/composition)
- Repeatability (less randomness)
- A workflow you can reuse
Entry: Image Creation (login required).
Start from templates (the most reliable way)
Templates are not a limitation—they’re a strong baseline for composition, lighting, camera language, and overall quality.
You can browse templates in two places:
Pick a template inside Image Creation
Open template search and choose one whose previews match your intent.
Don’t rewrite everything on the first try
A common beginner mistake is turning the prompt into an essay.
A more reliable workflow:
- Generate once to validate direction
- Then change one thing per iteration: expression / background / style / camera
When do you need reference images?
In Image Creation, you can upload up to 3 reference images. But note: not every model supports reference images.
Use this rule:
- If you need it to look like a specific person / product / subject, use reference images.
- If you only need a vibe (lighting/camera/style), start without references and use templates.
If you don’t see the reference upload UI, it usually means the current model doesn’t support reference images.
Options:
- Switch to a model that supports references (if available)
- Or rely on templates + text prompts
You can click the upload button to upload images.
You can also drag and drop images to upload.
We also support pasting images directly from the clipboard using Ctrl/Command + V.
Can you edit templates? Yes—here’s how to avoid making things worse
Yes, edit them—but do it with discipline:
- Start with the subject (who/what, action, mood)
- Then the frame (background, composition, camera)
- Only then add style words (realistic/film/anime/3D…)—don’t stack too many
A reliable iteration rule:
- Keep 80% of the template structure
- Change 20% (the one thing you actually want to improve)
Create your first reusable workflow in Image Creation
Start from templates, then iterate with small prompt changes.
Open Image CreationAfter generation: where to find history (and how to iterate)
Your generations live here: Generation History.
To turn “one lucky result into a repeatable workflow:
- Track: template, aspect ratio, key prompt lines
- Change one variable per iteration
- Generate 6–12 variations, then pick the best
Troubleshooting (check these first)
It’s close, but weird
- Your reference image lighting is too extreme—swap for a cleaner one
- The background is too busy—crop to isolate the subject
The subject is tiny / too much empty space
- Use a closer, portrait-style template
- Change aspect ratio (for portraits, start with
3:4or2:3)
The output looks flat / low-quality
- Reduce style-word stacking
- Describe lighting/camera/material first, then style
Prompts keep getting worse
- Revert to your last usable version
- Iterate by changing one thing at a time
FAQ
Should beginners start with AI Tools or Image Creation?
- Want fast results: start with AI Tools Hub
- Want control and repeatability: use Image Creation
Do I need to know prompting to use templates?
No. Templates are designed to package complex prompts into a reliable baseline.
When do I absolutely need reference images?
When identity or exact subject matching matters (a specific person/product), references usually improve stability.
Key Takeaways
- Start at AI Tools Hub for the fastest first win
- Use Image Creation for controllable, reusable creation
- Up to 3 reference images, depending on model support
- For consistency: change one variable per iteration