@imgly/plugin-ai-image-generation-web 0.2.6 → 0.2.8

package/CHANGELOG.md

@@ -2,6 +2,35 @@
 
 ## [Unreleased]
 
+## [0.2.8] - 2025-09-29
+
+### New Features
+
+-   [image-generation] **SeedreamV4 Provider**: Added ByteDance Seedream 4.0 text-to-image provider via fal.ai for high-quality image generation with multiple size presets (square HD 2048×2048, square 1024×1024, portrait/landscape variants), custom dimensions support (1024-4096 pixels), and safety checker enabled by default
+-   [image-generation] **SeedreamV4Edit Provider**: Added ByteDance Seedream 4.0 image-to-image provider via fal.ai for advanced image editing with unified generation/editing architecture, support for multiple input images (up to 10), and full canvas quick actions support (editImage, swapBackground, styleTransfer, artistTransfer, createVariant, combineImages, remixPage, remixPageWithPrompt)
+
+### Fixed
+
+-   [sticker-generation] **Fixed Input Types Not Enabled**: Fixed an issue where the sticker generation panel would show "No input types are enabled" error by properly enabling the `fromText` and `fromImage` feature flags during plugin initialization
+
+## [0.2.7] - 2025-09-26
+
+### New Features
+
+-   [image-generation] **QwenImageEdit Provider**: Added Qwen image editing provider via fal.ai for advanced image-to-image transformation with text prompts, supporting all standard quick actions
+-   [video-generation] **MinimaxHailuo02StandardImageToVideo Provider**: Added Minimax Hailuo-02 Standard image-to-video provider via fal.ai for transforming still images into videos with selectable resolutions (512P: 912×512, 768P: 1280×720) and adjustable durations (6 or 10 seconds)
+-   [video-generation] **ByteDance Seedance v1 Pro Providers**: Added ByteDance Seedance v1 Pro text-to-video and image-to-video providers via fal.ai with:
+    - Text-to-video generation from text descriptions with customizable aspect ratios
+    - Image-to-video transformation with dynamic motion generation from still images
+    - Multiple aspect ratio options (21:9, 16:9, 4:3, 1:1, 3:4, 9:16, or auto from image for i2v)
+    - Adjustable duration (3-12 seconds, default 5)
+    - Resolution options (480p, 720p, 1080p)
+    - Proper aspect ratio handling in placeholder blocks based on user selection
+
+-   [all] **Property Configuration System**: Providers can now define default values for their properties. Defaults can be static values or dynamic based on context (language, design state, etc.)
+
+-   [image-generation] **Recraft Provider Defaults**: Recraft providers (V3 and 20b) now support configurable default values for all properties, including dynamic style defaults based on the selected style type
+
 ## [0.2.6] - 2025-09-09
 
 ### New Features

package/README.md

@@ -176,6 +176,11 @@ text2image: FalAiImage.RecraftV3({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    style: { default: 'realistic_image' },  // Default style
+    image_size: { default: 'square_hd' }     // Default size
   }
 })
 ```
@@ -187,6 +192,97 @@ Key features:
 - Adjustable quality settings
 - Custom headers support for API requests
 
+**Available Property Values:**
+
+```typescript
+// image_size options:
+'square_hd'      // 1024x1024 (default)
+'square'         // 512x512
+'portrait_4_3'   // 768x1024
+'portrait_16_9'  // 576x1024
+'landscape_4_3'  // 1024x768
+'landscape_16_9' // 1024x576
+
+// style options (grouped by category):
+// Base styles:
+'any'                  // Let model decide
+'realistic_image'      // Photorealistic (default)
+'digital_illustration' // Digital art
+'vector_illustration'  // Vector graphics
+
+// Realistic substyles:
+'realistic_image/b_and_w'           // Black and white
+'realistic_image/hard_flash'        // Hard flash photography
+'realistic_image/hdr'               // HDR photography
+'realistic_image/natural_light'     // Natural lighting
+'realistic_image/studio_portrait'   // Studio portrait
+'realistic_image/enterprise'        // Business/corporate
+'realistic_image/motion_blur'       // Motion blur effect
+'realistic_image/evening_light'     // Evening lighting
+'realistic_image/faded_nostalgia'   // Nostalgic/faded look
+'realistic_image/forest_life'       // Forest/nature
+'realistic_image/mystic_naturalism' // Mystic nature
+'realistic_image/natural_tones'     // Natural color tones
+'realistic_image/organic_calm'      // Organic/calm aesthetic
+'realistic_image/real_life_glow'    // Natural glow
+'realistic_image/retro_realism'     // Retro realistic
+'realistic_image/retro_snapshot'    // Retro photo
+'realistic_image/urban_drama'       // Urban/dramatic
+'realistic_image/village_realism'   // Village/rural
+'realistic_image/warm_folk'         // Warm/folk style
+
+// Digital illustration substyles:
+'digital_illustration/pixel_art'           // Pixel art
+'digital_illustration/hand_drawn'          // Hand-drawn
+'digital_illustration/grain'               // Grainy texture
+'digital_illustration/infantile_sketch'    // Child-like sketch
+'digital_illustration/2d_art_poster'       // 2D poster art
+'digital_illustration/handmade_3d'         // Handmade 3D look
+'digital_illustration/hand_drawn_outline'  // Hand-drawn outline
+'digital_illustration/engraving_color'     // Color engraving
+'digital_illustration/2d_art_poster_2'     // Alternative poster style
+'digital_illustration/antiquarian'         // Antique/vintage
+'digital_illustration/bold_fantasy'        // Bold fantasy art
+'digital_illustration/child_book'          // Children's book
+'digital_illustration/child_books'         // Children's book alt
+'digital_illustration/cover'               // Book/album cover
+'digital_illustration/crosshatch'          // Crosshatching
+'digital_illustration/digital_engraving'   // Digital engraving
+'digital_illustration/expressionism'       // Expressionist
+// ... and many more digital illustration styles
+
+// Vector illustration substyles:
+'vector_illustration/cartoon'              // Cartoon style
+'vector_illustration/kawaii'               // Kawaii/cute
+'vector_illustration/comic'                // Comic book
+'vector_illustration/line_art'             // Line art
+'vector_illustration/noir_silhouette'      // Noir silhouette
+```
+
+**Configuration Example with All Properties:**
+
+```typescript
+text2image: FalAiImage.RecraftV3({
+  proxyUrl: 'http://your-proxy-server.com/api/proxy',
+  properties: {
+    prompt: { default: '' },  // User will fill this
+    style: { default: 'realistic_image/natural_light' },
+    image_size: { default: 'landscape_16_9' },
+
+    // Dynamic style based on context
+    style: {
+      default: (context) => {
+        const hour = new Date().getHours();
+        if (hour < 6 || hour > 18) {
+          return 'realistic_image/evening_light';
+        }
+        return 'realistic_image/natural_light';
+      }
+    }
+  }
+})
+```
+
 **Style Group Control:**
 You can control which style groups (image/vector) are available using the Feature API:
 
@@ -221,6 +317,19 @@ text2image: FalAiImage.Recraft20b({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    // Dynamic style defaults based on style type
+    style: {
+      default: (context) => {
+        // Different defaults for different style categories
+        // This is handled internally by the provider
+        return 'broken_line';  // Default for icon styles
+      }
+    },
+    image_size: { default: 'square_hd' },
+    n_colors: { default: 4 }  // Default color count for icon styles
   }
 })
 ```
@@ -232,6 +341,67 @@ Key features:
 - Same image size presets and custom dimensions support
 - Cost-effective alternative to RecraftV3
 
+**Available Property Values:**
+
+```typescript
+// image_size options (same as RecraftV3):
+'square_hd'      // 1024x1024 (default)
+'square'         // 512x512
+'portrait_4_3'   // 768x1024
+'portrait_16_9'  // 576x1024
+'landscape_4_3'  // 1024x768
+'landscape_16_9' // 1024x576
+
+// style options - includes all RecraftV3 styles PLUS icon styles:
+// All RecraftV3 styles are supported (see RecraftV3 above)
+
+// Additional Icon styles (unique to Recraft20b):
+'icon/broken_line'            // Broken line icon style
+'icon/colored_outline'        // Colored outline icons
+'icon/colored_shapes'         // Solid colored shapes
+'icon/colored_shapes_gradient' // Gradient colored shapes
+'icon/doodle_fill'           // Doodle-filled icons
+'icon/doodle_offset_fill'    // Offset doodle fill
+'icon/offset_fill'           // Offset fill style
+'icon/outline'               // Simple outline
+'icon/outline_gradient'      // Gradient outline
+'icon/uneven_fill'          // Uneven/artistic fill
+
+// Logo styles:
+'logo'                       // Logo design
+
+// n_colors options (for icon styles):
+1  // Monochrome
+2  // Two colors (default)
+3  // Three colors
+4  // Four colors
+// ... up to any reasonable number
+```
+
+**Configuration Example with All Properties:**
+
+```typescript
+text2image: FalAiImage.Recraft20b({
+  proxyUrl: 'http://your-proxy-server.com/api/proxy',
+  properties: {
+    prompt: { default: '' },  // User will fill this
+    style: { default: 'icon/colored_outline' },  // Default to icon style
+    image_size: { default: 'square' },
+    n_colors: { default: 3 },  // Number of colors for icon styles
+
+    // Dynamic style based on use case
+    style: {
+      default: (context) => {
+        // You could check block type or other context
+        const engine = context.engine;
+        // Return appropriate style
+        return 'icon/broken_line';
+      }
+    }
+  }
+})
+```
+
 **Style Group Control:**
 You can control which style groups (image/vector/icon) are available using the Feature API:
 
@@ -270,6 +440,12 @@ text2image: OpenAiImage.GptImage1.Text2Image({
   headers: {
     'x-api-key': 'your-key',
     'x-request-source': 'cesdk-plugin'
+  },
+  // Optional: Configure default property values
+  properties: {
+    size: { default: '1024x1024' },  // Options: '1024x1024', '1792x1024', '1024x1792'
+    quality: { default: 'standard' }, // Options: 'standard', 'hd'
+    style: { default: 'vivid' }      // Options: 'vivid', 'natural'
   }
 })
 ```
@@ -291,6 +467,12 @@ image2image: FalAiImage.GeminiFlashEdit({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    strength: { default: 0.8 },        // Transformation strength (0.0-1.0)
+    guidance_scale: { default: 7.5 },  // Guidance scale (0-20)
+    num_inference_steps: { default: 50 }  // Number of inference steps
   }
 })
 ```
@@ -343,6 +525,12 @@ text2image: FalAiImage.IdeogramV3({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    style: { default: 'GENERAL' },    // Options: 'AUTO', 'GENERAL', 'REALISTIC', 'DESIGN'
+    image_size: { default: 'square_hd' }, // Same options as Recraft
+    seed: { default: 12345 }         // Fixed seed for reproducibility
   }
 })
 ```
@@ -362,6 +550,12 @@ image2image: FalAiImage.IdeogramV3Remix({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    style: { default: 'AUTO' },       // Options: 'AUTO', 'GENERAL', 'REALISTIC', 'DESIGN'
+    image_size: { default: 'square_hd' },
+    remix_strength: { default: 0.7 }  // How much to transform (0.0-1.0)
   }
 })
 ```
@@ -372,7 +566,34 @@ Key features:
 - Multiple image size presets: square HD, square, portrait 4:3/16:9, landscape 4:3/16:9
 - Custom dimensions support (64x64 to 14142x14142 pixels)
 
-#### 7. FluxProKontextEdit (Image-to-Image)
+#### 7. QwenImageEdit (Image-to-Image)
+
+A powerful image editing model with superior text editing capabilities from fal.ai:
+
+```typescript
+image2image: FalAiImage.QwenImageEdit({
+  proxyUrl: 'http://your-proxy-server.com/api/proxy',
+  headers: {
+    'x-custom-header': 'value',
+    'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    seed: { default: 12345 },         // Fixed seed for reproducibility
+    guidance_scale: { default: 7.5 }  // Guidance strength (0-20)
+  }
+})
+```
+
+Key features:
+- **Superior text editing capabilities** for image transformation
+- Transform existing images with detailed text prompts
+- Available through all canvas quick actions (edit, background swap, style transfer, artist styles, variants)
+- Maintains original image dimensions
+- Custom headers support for API requests
+```
+
+#### 8. FluxProKontextEdit (Image-to-Image)
 
 A versatile image editing model that applies stylistic changes and transformations:
 
@@ -387,7 +608,7 @@ Key features:
 - Built-in quick actions for style transfer, artist styles, background swapping, and variants
 - Keeps original image dimensions
 
-#### 8. FluxProKontextMaxEdit (Image-to-Image)
+#### 9. FluxProKontextMaxEdit (Image-to-Image)
 
 A high-quality variant of FluxProKontext offering more detailed results:
 
@@ -413,6 +634,10 @@ text2image: FalAiImage.NanoBanana({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    output_format: { default: 'PNG' }  // Options: 'PNG', 'JPEG'
   }
 })
 ```
@@ -437,7 +662,53 @@ cesdk.i18n.setTranslations({
 });
 ```
 
-#### 10. NanoBananaEdit (Image-to-Image)
+#### 10. SeedreamV4 (Text-to-Image)
+
+A powerful text-to-image model from ByteDance's Seedream 4.0 available through fal.ai:
+
+```typescript
+text2image: FalAiImage.SeedreamV4({
+  proxyUrl: 'http://your-proxy-server.com/api/proxy',
+  headers: {
+    'x-custom-header': 'value',
+    'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    image_size: { default: 'square_hd' }  // Options: square_hd, square, portrait/landscape variants
+  }
+})
+```
+
+Key features:
+- High-quality image generation with ByteDance's Seedream 4.0 model
+- Multiple image size presets: square HD (2048×2048), square (1024×1024), portrait 4:3/16:9, landscape 4:3/16:9
+- Custom dimensions support (1024-4096 pixels)
+- Fast generation times
+- Safety checker enabled by default
+
+#### 11. SeedreamV4Edit (Image-to-Image)
+
+An advanced image editing model from ByteDance's Seedream 4.0 for transforming existing images:
+
+```typescript
+image2image: FalAiImage.SeedreamV4Edit({
+  proxyUrl: 'http://your-proxy-server.com/api/proxy',
+  headers: {
+    'x-custom-header': 'value',
+    'x-client-version': '1.0.0'
+  }
+})
+```
+
+Key features:
+- Unified architecture for both generation and editing
+- Supports multiple input images (up to 10)
+- Full canvas quick actions support: edit image, swap background, style transfer, artist styles, create variants, combine images, remix page
+- Maintains original image dimensions
+- Custom headers support for API requests
+
+#### 12. NanoBananaEdit (Image-to-Image)
 
 An image editing model from fal.ai that transforms existing images using text prompts:
 
@@ -447,6 +718,11 @@ image2image: FalAiImage.NanoBananaEdit({
   headers: {
     'x-custom-header': 'value',
     'x-client-version': '1.0.0'
+  },
+  // Optional: Configure default property values
+  properties: {
+    strength: { default: 0.7 },        // Edit strength (0.0-1.0)
+    output_format: { default: 'PNG' }  // Options: 'PNG', 'JPEG'
   }
 })
 ```
@@ -795,6 +1071,16 @@ FalAiImage.GeminiFlashEdit(config: {
 })
 ```
 
+#### QwenImageEdit
+
+```typescript
+FalAiImage.QwenImageEdit(config: {
+  proxyUrl: string;
+  headers?: Record<string, string>;
+  debug?: boolean;
+})
+```
+
 ### OpenAI Providers
 
 #### GptImage1.Text2Image
@@ -855,6 +1141,26 @@ FalAiImage.NanoBananaEdit(config: {
 })
 ```
 
+#### SeedreamV4
+
+```typescript
+FalAiImage.SeedreamV4(config: {
+  proxyUrl: string;
+  headers?: Record<string, string>;
+  debug?: boolean;
+})
+```
+
+#### SeedreamV4Edit
+
+```typescript
+FalAiImage.SeedreamV4Edit(config: {
+  proxyUrl: string;
+  headers?: Record<string, string>;
+  debug?: boolean;
+})
+```
+
 ## UI Integration
 
 The plugin automatically registers the following UI components:
@@ -943,12 +1249,15 @@ const myImageProvider = {
   - IdeogramV3: `ly.img.ai.fal-ai/ideogram/v3`
   - IdeogramV3Remix: `ly.img.ai.fal-ai/ideogram/v3/remix`
   - GeminiFlashEdit: `ly.img.ai.fal-ai/gemini-flash-edit`
+  - QwenImageEdit: `ly.img.ai.fal-ai/qwen-image-edit`
   - GptImage1.Text2Image: `ly.img.ai.open-ai/gpt-image-1/text2image`
   - GptImage1.Image2Image: `ly.img.ai.open-ai/gpt-image-1/image2image`
   - FluxProKontextEdit: `ly.img.ai.fal-ai/flux-pro/kontext`
   - FluxProKontextMaxEdit: `ly.img.ai.fal-ai/flux-pro/kontext/max`
   - NanoBanana: `ly.img.ai.fal-ai/nano-banana`
   - NanoBananaEdit: `ly.img.ai.fal-ai/nano-banana/edit`
+  - SeedreamV4: `ly.img.ai.fal-ai/bytedance/seedream/v4/text-to-image`
+  - SeedreamV4Edit: `ly.img.ai.fal-ai/bytedance/seedream/v4/edit`
 
 ### Asset History
 
@@ -958,12 +1267,15 @@ Generated images are automatically stored in asset sources with the following ID
 - IdeogramV3: `fal-ai/ideogram/v3.history`
 - IdeogramV3Remix: `fal-ai/ideogram/v3/remix.history`
 - GeminiFlashEdit: `fal-ai/gemini-flash-edit.history`
+- QwenImageEdit: `fal-ai/qwen-image-edit.history`
 - GptImage1.Text2Image: `open-ai/gpt-image-1/text2image.history`
 - GptImage1.Image2Image: `open-ai/gpt-image-1/image2image.history`
 - FluxProKontextEdit: `fal-ai/flux-pro/kontext.history`
 - FluxProKontextMaxEdit: `fal-ai/flux-pro/kontext/max.history`
 - NanoBanana: `fal-ai/nano-banana.history`
 - NanoBananaEdit: `fal-ai/nano-banana/edit.history`
+- SeedreamV4: `fal-ai/bytedance/seedream/v4/text-to-image.history`
+- SeedreamV4Edit: `fal-ai/bytedance/seedream/v4/edit.history`
 
 ### Dock Integration