techfreakworm commited on
Commit
3aed2df
·
unverified ·
1 Parent(s): 7953225

feat(t2i): raise Width/Height cap to 2048 and switch step to 16

Browse files

Z-Image-Base's model card explicitly documents the supported range as
"512x512 to 2048x2048 (total pixel area, any aspect ratio)". The previous
cap of 1536 with step=64 clipped valid configurations and contradicted
the actual mod-16 constraint (8x VAE + 2x patch).

- Width / Height sliders: 384..1536 step 64 -> 384..2048 step 16.
- Tooltips: "Multiples of 64" was wrong; correct constraint is 16. Call
out the 1024-1280 sweet spot and the documented 2048x2048 area max so
users know what's supported vs untrained territory.

Turbo is a distillation of Base and inherits the same dim constraints.

Files changed (2) hide show
  1. tooltips.py +2 -2
  2. ui.py +2 -2
tooltips.py CHANGED
@@ -14,8 +14,8 @@ TOOLTIPS: dict[str, str] = {
14
  "lora_strength": "LoRA influence. 0.6-1.0 typical. Higher = more LoRA, less base model.",
15
  "steps": "Denoising steps. Turbo: 6-10. Base: 20-30. More = better detail, slower.",
16
  "cfg": "Classifier-free guidance. Turbo: locked at 1.0. Base: 3-5 typical.",
17
- "width": "Output width in pixels. Multiples of 64. Higher = more memory.",
18
- "height": "Output height in pixels. Multiples of 64.",
19
  "seed": "0 = random each run. Pin a number to reproduce an image exactly.",
20
  "controlnet_image": "Control image — the structural reference for the output.",
21
  "controlnet_preprocessor": "Canny = edges, Depth = depth map, Pose = body pose, Pre-processed = use image as-is.",
 
14
  "lora_strength": "LoRA influence. 0.6-1.0 typical. Higher = more LoRA, less base model.",
15
  "steps": "Denoising steps. Turbo: 6-10. Base: 20-30. More = better detail, slower.",
16
  "cfg": "Classifier-free guidance. Turbo: locked at 1.0. Base: 3-5 typical.",
17
+ "width": "Output width in pixels. Multiples of 16. Sweet spot 1024-1280; supported up to 2048x2048 total area.",
18
+ "height": "Output height in pixels. Multiples of 16. Sweet spot 1024-1280; supported up to 2048x2048 total area.",
19
  "seed": "0 = random each run. Pin a number to reproduce an image exactly.",
20
  "controlnet_image": "Control image — the structural reference for the output.",
21
  "controlnet_preprocessor": "Canny = edges, Depth = depth map, Pose = body pose, Pre-processed = use image as-is.",
ui.py CHANGED
@@ -89,8 +89,8 @@ def build_t2i_tab() -> dict[str, gr.components.Component]:
89
  steps = gr.Slider(1, 50, value=8, step=1, label="Steps", info=TOOLTIPS["steps"])
90
 
91
  with gr.Accordion("Advanced", open=False):
92
- width = gr.Slider(384, 1536, value=1024, step=64, label="Width", info=TOOLTIPS["width"])
93
- height = gr.Slider(384, 1536, value=1024, step=64, label="Height", info=TOOLTIPS["height"])
94
  seed = gr.Number(value=0, precision=0, label="Seed", info=TOOLTIPS["seed"])
95
 
96
  generate_btn = gr.Button("Generate", variant="primary")
 
89
  steps = gr.Slider(1, 50, value=8, step=1, label="Steps", info=TOOLTIPS["steps"])
90
 
91
  with gr.Accordion("Advanced", open=False):
92
+ width = gr.Slider(384, 2048, value=1024, step=16, label="Width", info=TOOLTIPS["width"])
93
+ height = gr.Slider(384, 2048, value=1024, step=16, label="Height", info=TOOLTIPS["height"])
94
  seed = gr.Number(value=0, precision=0, label="Seed", info=TOOLTIPS["seed"])
95
 
96
  generate_btn = gr.Button("Generate", variant="primary")