⟡ img.lab
Prototype

Image lab workspace

Compose prompts, set guidance, and preview requests before they hit the selected model. This scaffolding will connect to the generation stack once the LLM + diffusion choice is set.

Open workbench Admin panel
Prompt

Describe the image

Ready Text to Image

Steps: (1) Pick a model (Flux Schnell = faster, SDXL = highest realism). (2) Write a clear prompt. (3) For image-to-image, upload a source image and switch to the Image tab.

Describe the subject, lighting, style, camera, background.
Tip: Flux Schnell for speed; SDXL for highest quality.

Estimated cost: $0.0019 (512×512)

Use one or more images; first image is sent. Switch to Text tab to ignore uploads.
Advanced: resolution
Aspect selection coming soon
Admin panel

Estimated cost: $0.0019 (512×512)

Preview

Output slot

Idle
Submit a prompt to generate Result will appear here
Activity
Local only
  • Idle
  • Default model: realistic_vision_1_3 ($0.0019 (512×512))
  • Mode: Text to Image
Models

Choose the right model

Dezgo

A quick guide to common options. Higher resolution or specialty models may cost more (see pricing in admin).

sd1
Sample
Realistic Vision 1.3
realistic_vision_1_3
Native: 512×512 $0.0019 (512×512)
Realistic portraits/people at 512px.

Triggers: RAW photo, *subject*, high detailed skin, 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3

sdxl
Sample coming soon
BluePencilXL 3.1.0
bluepencilxl_1024px
Native: 1024×1024 $0.0029 (1024×1024)
High-res anime style.
sdxl_lightning
Sample coming soon
DreamshaperXL Lightning
dreamshaperxl_lightning_1024px
Native: 1024×1024 $0.0032 (1024×1024)
High-quality generalist; great realism at 1024px.
flux_schnell
Sample coming soon
Flux Schnell
flux_1_schnell
Native: 512×512 $0.0019 (512×512)
Fast, cutting-edge quality and strong prompt following. Good default.