test: empty signed commit to verify HF GPG verification

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

refactor(upscale): drop LoRA support per spec

fix(controlnet): crop control image to multiple-of-16 dims

fix(lora): accept diffusion_model.* prefix and use pipe.load_lora hotload

fix(upscale): crop upscaled image to multiple-of-16 dims before refinement

docs: polish README + write AGENTS.md + SKILLS.md, refresh CLAUDE.md

feat(ui): support CTA bar — '♥ at the top to support · follow for what's next'

test: update _on_model_change asserts for new 3-tuple return shape

fix(upscale): pass post-upscale dims to pipe + LoRA label hints model compat

fix: depth preprocessor name + basicsr/torchvision import shim

perf(preload): broaden globs to capture configs + index.jsons (~47 GB total)

chore(ui): display brand as 'Z-Image Studio' (title-case) in header + browser tab

feat(ui): soft dark restraint redesign — native gr.Radio + accordion + lora toggle

fix: actually use Base transformer when user picks Base

feat(progress): gradio progress(track_tqdm=True) on all three handlers

fix: pool-stashed transformer swap + MPS-safe vram + corrected model-zoo anchor

fix(deps): add torchaudio + pin transformers<5

fix(deploy): switch model lookups to HF + fix ZeroGPU build

fix(realesrgan): switch repo from xinntao/Real-ESRGAN to lllyasviel/Annotators

fix(deps): install diffsynth from github (not on pypi)

fix: post-review polish — slider defaults, error handling, comment

test: l3 gpu smoke (t2i base/turbo + controlnet + upscale)

ci: ruff + pytest on push/pr (l1+l2, no gpu deps)

docs: hf space frontmatter + readme

feat(app): gradio blocks entrypoint with bootstrap + event wiring + js shim

feat(ui): per-tab gradio builders with labeled_label + custom model selector

feat(backend): ZImageStudioBackend with @spaces.GPU and mode dispatch

feat(backend): zerogpu duration estimator (clamped 60-180s)

feat(modes): upscale handler (realesrgan + z-image-turbo refinement)

feat(modes): controlnet handler (turbo + union 2.1 + preprocessor)

feat(modes): t2i handler (base + turbo) with transformer swap and lora ctx

feat(upscale): realesrgan x4 wrapper with 0.5-resize bridge

feat(preprocessors): canny/depth/pose via controlnet_aux (lazy imports)

feat(lora): applied_lora ctx manager — validate, apply, revert on exit

feat(lora): safetensors header sniff + zimage key validation

feat(models): hf cache mirror (hardlink blobs, preserve snapshot symlinks, copy refs)

feat(ui): labeled_label and model_selector_html helpers

feat(theme): css rules for param info tooltips and custom model selector

refactor(tooltips): rename copy.py → tooltips.py to avoid stdlib shadow

feat(copy): tooltip strings dict — one source of truth for param descriptions

docs: amend spec + plan with param tooltips and coming-soon model placeholders

feat(models): device autodetect, vram-limit helpers, model config registry

fix(theme): preserve gradio's --font / --font-mono css vars

feat(theme): onyx amber palette + gr.themes.Base + glow CSS

chore: project scaffolding (pyproject, requirements, license, claude.md, tests)

docs: implementation plan for z-image-studio (19 tasks)

docs: initial design spec for z-image-studio