| name | photo-studio |
| description | Local AI photo generation and editing using FLUX.2 Klein 4B and Z-Image Turbo. Use when the user wants to generate or edit images locally (no API cost, no rate limits). Models at I:/AI-Models/. Script at G:/Projects/Kolbo.AI/github/training-loras/scripts/photo-studio.py Keywords: generate image, edit image, flux klein, z-image turbo, local diffusion, photo studio
|
Photo Studio — Local AI Image Generation & Editing
Infrastructure
| Component | Path |
|---|
| Script | G:/Projects/Kolbo.AI/github/training-loras/scripts/photo-studio.py |
| Python venv | G:/Projects/Kolbo.AI/github/ai-toolkit/venv/Scripts/python.exe |
| FLUX.2 Klein 4B | I:/AI-Models/flux2-klein-4b/ |
| Z-Image Turbo | I:/AI-Models/z-image-turbo/ |
| Z-Image Adapter | I:/AI-Models/z-image-turbo-adapter/zimage_turbo_training_adapter_v2.safetensors |
| Vision / LLM | Agent's built-in vision (Read tool) for images. For video analysis load the video-production skill. |
How to Run
Always use the ai-toolkit venv (has Flux2KleinPipeline + ZImagePipeline):
"G:/Projects/Kolbo.AI/github/ai-toolkit/venv/Scripts/python.exe" \
"G:/Projects/Kolbo.AI/github/training-loras/scripts/photo-studio.py" \
[args...]
CLI Reference
| Flag | Default | Description |
|---|
--prompt "..." | — | Text prompt |
--image path.jpg | — | Input image (FLUX editing only) |
--model flux|zimage | flux | Model to use |
--output result.jpg | output_<ts>.jpg | Output path |
--width N | 1152 | Width in pixels |
--height N | 2048 | Height in pixels |
--steps N | 20 | Inference steps |
--cfg N | 3.5 | Guidance scale |
--seed N | random | Deterministic seed |
--adapter | off | Load Z-Image Turbo adapter (zimage only) |
Common Recipes
Text-to-image (portrait)
python photo-studio.py \
--prompt "full-body front view, woman in designer dress, cinematic lighting" \
--model flux --width 1152 --height 2048
Edit existing image
python photo-studio.py \
--image input.jpg \
--prompt "wide full-body view of the subject, cinematic" \
--model flux --width 1152 --height 2048
Analyze image then generate variation
# Step 1: Read the image — the agent sees it natively (built-in vision)
Read("/abs/path/to/char.jpg")
→ Agent describes the person: clothing, pose, features, style
# Step 2: Use the description as the prompt
python photo-studio.py \
--prompt "<description from agent vision> standing upright, full body" \
--model flux
Enhance a short prompt then generate
# Step 1: Enhance the prompt with Kolbo MCP
chat_send_message({
message: "Expand this into a detailed image generation prompt for a photorealistic portrait: 'street fashion guy'",
})
→ { content: "A young man in his mid-20s wearing..." }
# Step 2: Generate with the enhanced prompt
python photo-studio.py \
--prompt "<enhanced prompt from Kolbo>" \
--model zimage --width 1152 --height 2048
Z-Image with adapter (MIRAGE-style)
python photo-studio.py \
--prompt "cinematic full-body portrait, editorial" \
--model zimage --adapter --width 1152 --height 1536 --cfg 4.0
Model Notes
FLUX.2 Klein 4B (--model flux)
- Class:
Flux2KleinPipeline (diffusers)
- Supports both text-to-image AND image editing via
image param
- Good default:
--steps 20 --cfg 3.5
- ~13GB VRAM needed
Z-Image Turbo (--model zimage)
- Class:
ZImagePipeline (diffusers), uses Qwen3 text encoder
- Text-to-image ONLY (no image editing)
- 8 NFEs, no CFG — distilled model, per official README. More steps = waste, CFG > 0 = wrong
- Default:
--steps 8 --cfg 0.0 --width 1152 --height 2048 (~30s per image)
- Add
--adapter to load the v2 training adapter
Vision & Prompt Enhancement
- For image analysis: use the agent's built-in vision —
Read the image file directly, no MCP needed
- For prompt enhancement:
chat_send_message asking Kolbo to expand a short prompt (text-only, no vision)
- Do NOT use
--analyze or --enhance flags (those call a local model that is no longer used)
When to use which model
| Task | Model |
|---|
| Edit an existing image | flux |
| Character body panel generation | flux |
| MIRAGE-style portraits | zimage --adapter |
| Quick text-to-image | flux or zimage |
| Portrait + face reference | flux --image face.jpg |
Prompt Tips
When the user gives a short/vague prompt, use chat_send_message to let Kolbo AI expand it before passing to the script. For image editing, first analyze the source image with the agent's built-in vision (Read the image), then use the description as the base prompt.
