mirror of https://github.com/Comfy-Org/ComfyUI_frontend.git synced 2026-04-20 14:30:41 +00:00

Files

dante01yoon bbd0a6b201 feat: migrate workflow template site as apps/hub

Migrate workflow_templates/site into the frontend monorepo as apps/hub
so the hub can use @comfyorg/design-system and shared packages.

Changes to existing files:
- pnpm-workspace.yaml: add @astrojs/sitemap, @astrojs/vercel, lucide-vue-next
- eslint.config.ts: add hub ignores and i18n/import rule overrides
- .oxlintrc.json: add hub scripts to ignore patterns
- knip.config.ts: add hub workspace config

apps/hub adaptations from source:
- Replace local cn() with @comfyorg/tailwind-utils (19 files)
- Integrate @comfyorg/design-system/css/base.css in global.css
- Make TEMPLATES_DIR configurable via HUB_TEMPLATES_DIR env var
- Add HUB_SKIP_SYNC flag for builds without template data
- Remove Vite 8-incompatible rollupOptions.output.manualChunks
- Fix stylelint violations (modern color notation, number precision)
- Gitignore generated content (thumbnails, synced templates, AI cache)

2026-04-06 20:53:13 +09:00

1.4 KiB

Raw Blame History

Z-Image

Z-Image is Zhipu AI's image generation model family, built on the CogView architecture with a hybrid autoregressive and diffusion decoder design.

Model Variants

GLM-Image (Z-Image)

9B autoregressive + 7B DiT diffusion decoder hybrid architecture
First open-source industrial-grade discrete autoregressive image generator
State-of-the-art bilingual text rendering (English and Chinese)

Z-Image-Turbo

Optimized variant for faster inference with reduced latency
Suitable for real-time and batch generation workflows

CogView-4

6B parameter DiT diffusion model, foundation for the Z-Image decoder

Key Features

Industry-leading text rendering accuracy for posters and infographics
Custom resolution from 512px to 2048px (multiples of 32)
Image editing, style transfer, and identity-preserving generation
LoRA training support; open weights on HuggingFace

Hardware Requirements

Cloud API: no local hardware required ($0.015 per image via Z.ai)
Self-hosted: 24GB+ VRAM for the combined 9B+7B architecture

Common Use Cases

Text-to-image generation with accurate text rendering
Commercial poster and graphic design
Social media content creation
Multi-subject consistency and identity-preserving generation

Key Parameters

prompt: text description of the desired image
size: output resolution (e.g., 1280x1280, 1568x1056, 960x1728)
model: glm-image or cogview-4

1.4 KiB Raw Blame History