@weirdfingers/baseboards 0.6.1 → 0.7.0

package/templates/api/src/boards/generators/implementations/fal/audio/chatterbox_tts_turbo.py

@@ -0,0 +1,195 @@
+"""
+fal.ai Chatterbox Text-to-Speech Turbo generator.
+
+Generate expressive speech from text with paralinguistic controls like laughs,
+sighs, coughs, and more. Supports voice cloning with custom audio samples.
+
+Based on Fal AI's fal-ai/chatterbox/text-to-speech/turbo model.
+See: https://fal.ai/models/fal-ai/chatterbox/text-to-speech/turbo
+"""
+
+import os
+from typing import Literal
+
+from pydantic import BaseModel, Field
+
+from ....artifacts import AudioArtifact
+from ....base import BaseGenerator, GeneratorExecutionContext, GeneratorResult
+
+# Voice presets available in Chatterbox
+ChatterboxVoice = Literal[
+    "aaron",
+    "abigail",
+    "anaya",
+    "andy",
+    "archer",
+    "brian",
+    "chloe",
+    "dylan",
+    "emmanuel",
+    "ethan",
+    "evelyn",
+    "gavin",
+    "gordon",
+    "ivan",
+    "laura",
+    "lucy",
+    "madison",
+    "marisol",
+    "meera",
+    "walter",
+]
+
+
+class ChatterboxTtsTurboInput(BaseModel):
+    """Input schema for Chatterbox TTS Turbo.
+
+    Artifact fields are automatically detected via type introspection
+    and resolved from generation IDs to artifact objects.
+    """
+
+    text: str = Field(
+        description=(
+            "The text to be converted to speech. Supports paralinguistic tags: "
+            "[clear throat], [sigh], [shush], [cough], [groan], [sniff], [gasp], "
+            "[chuckle], [laugh]"
+        ),
+        min_length=1,
+    )
+    voice: ChatterboxVoice = Field(
+        default="lucy",
+        description="Preset voice to use for synthesis",
+    )
+    audio_url: AudioArtifact | None = Field(
+        default=None,
+        description=(
+            "Optional audio file (5-10 seconds) for voice cloning. "
+            "If provided, this overrides the preset voice selection."
+        ),
+    )
+    temperature: float = Field(
+        default=0.8,
+        ge=0.05,
+        le=2.0,
+        description="Temperature for generation. Higher values create more varied speech patterns.",
+    )
+    seed: int | None = Field(
+        default=None,
+        description="Random seed for reproducible results. Set for consistent generations.",
+    )
+
+
+class FalChatterboxTtsTurboGenerator(BaseGenerator):
+    """Chatterbox TTS Turbo text-to-speech generator using fal.ai."""
+
+    name = "fal-chatterbox-tts-turbo"
+    artifact_type = "audio"
+    description = (
+        "Fal: Chatterbox TTS Turbo - "
+        "Expressive text-to-speech with paralinguistic controls and voice cloning"
+    )
+
+    def get_input_schema(self) -> type[ChatterboxTtsTurboInput]:
+        return ChatterboxTtsTurboInput
+
+    async def generate(
+        self, inputs: ChatterboxTtsTurboInput, context: GeneratorExecutionContext
+    ) -> GeneratorResult:
+        """Generate audio using fal.ai Chatterbox TTS Turbo model."""
+        # Check for API key (fal-client uses FAL_KEY environment variable)
+        if not os.getenv("FAL_KEY"):
+            raise ValueError("API configuration invalid. Missing FAL_KEY environment variable")
+
+        # Import fal_client
+        try:
+            import fal_client
+        except ImportError as e:
+            raise ImportError(
+                "fal.ai SDK is required for FalChatterboxTtsTurboGenerator. "
+                "Install with: pip install weirdfingers-boards[generators-fal]"
+            ) from e
+
+        # Prepare arguments for fal.ai API
+        arguments: dict[str, str | float | int] = {
+            "text": inputs.text,
+            "voice": inputs.voice,
+            "temperature": inputs.temperature,
+        }
+
+        # Add seed if provided
+        if inputs.seed is not None:
+            arguments["seed"] = inputs.seed
+
+        # Handle voice cloning audio upload
+        if inputs.audio_url is not None:
+            from ..utils import upload_artifacts_to_fal
+
+            audio_urls = await upload_artifacts_to_fal([inputs.audio_url], context)
+            arguments["audio_url"] = audio_urls[0]
+
+        # Submit async job and get handler
+        handler = await fal_client.submit_async(
+            "fal-ai/chatterbox/text-to-speech/turbo",
+            arguments=arguments,
+        )
+
+        # Store the external job ID for tracking
+        await context.set_external_job_id(handler.request_id)
+
+        # Stream progress updates (sample every 3rd event to avoid spam)
+        from .....progress.models import ProgressUpdate
+
+        event_count = 0
+        async for event in handler.iter_events(with_logs=True):
+            event_count += 1
+
+            # Process every 3rd event to provide feedback without overwhelming
+            if event_count % 3 == 0:
+                # Extract logs if available
+                logs = getattr(event, "logs", None)
+                if logs:
+                    # Join log entries into a single message
+                    if isinstance(logs, list):
+                        message = " | ".join(str(log) for log in logs if log)
+                    else:
+                        message = str(logs)
+
+                    if message:
+                        await context.publish_progress(
+                            ProgressUpdate(
+                                job_id=handler.request_id,
+                                status="processing",
+                                progress=50.0,  # Approximate mid-point progress
+                                phase="processing",
+                                message=message,
+                            )
+                        )
+
+        # Get final result
+        result = await handler.get()
+
+        # Extract audio URL from result
+        # fal.ai returns: {"audio": {"url": "..."}}
+        audio_data = result.get("audio")
+        if audio_data is None:
+            raise ValueError("No audio data returned from fal.ai API")
+
+        audio_url = audio_data.get("url")
+        if not audio_url:
+            raise ValueError("Audio URL missing in fal.ai response")
+
+        # Store audio result
+        artifact = await context.store_audio_result(
+            storage_url=audio_url,
+            format="wav",  # Chatterbox TTS returns WAV format
+            output_index=0,
+        )
+
+        return GeneratorResult(outputs=[artifact])
+
+    async def estimate_cost(self, inputs: ChatterboxTtsTurboInput) -> float:
+        """Estimate cost for Chatterbox TTS Turbo generation.
+
+        Chatterbox TTS Turbo pricing is approximately $0.03 per generation.
+        """
+        return 0.03

package/templates/api/src/boards/generators/implementations/fal/image/__init__.py

@@ -1,5 +1,6 @@
 """Fal.ai image generators."""
 
+from .bytedance_seedream_v45_edit import FalBytedanceSeedreamV45EditGenerator
 from .clarity_upscaler import FalClarityUpscalerGenerator
 from .crystal_upscaler import FalCrystalUpscalerGenerator
 from .fal_ideogram_character import FalIdeogramCharacterGenerator
@@ -10,8 +11,11 @@ from .flux_2_pro_edit import FalFlux2ProEditGenerator
 from .flux_pro_kontext import FalFluxProKontextGenerator
 from .flux_pro_ultra import FalFluxProUltraGenerator
 from .gemini_25_flash_image import FalGemini25FlashImageGenerator
+from .gemini_25_flash_image_edit import FalGemini25FlashImageEditGenerator
+from .gpt_image_1_5 import FalGptImage15Generator
 from .gpt_image_1_edit_image import FalGptImage1EditImageGenerator
 from .gpt_image_1_mini import FalGptImage1MiniGenerator
+from .gpt_image_15_edit import FalGptImage15EditGenerator
 from .ideogram_character_edit import FalIdeogramCharacterEditGenerator
 from .ideogram_v2 import FalIdeogramV2Generator
 from .imagen4_preview import FalImagen4PreviewGenerator
@@ -22,8 +26,12 @@ from .nano_banana_pro import FalNanoBananaProGenerator
 from .nano_banana_pro_edit import FalNanoBananaProEditGenerator
 from .qwen_image import FalQwenImageGenerator
 from .qwen_image_edit import FalQwenImageEditGenerator
+from .reve_edit import FalReveEditGenerator
+from .reve_text_to_image import FalReveTextToImageGenerator
+from .seedream_v45_text_to_image import FalSeedreamV45TextToImageGenerator
 
 __all__ = [
+    "FalBytedanceSeedreamV45EditGenerator",
     "FalClarityUpscalerGenerator",
     "FalCrystalUpscalerGenerator",
     "FalFlux2Generator",
@@ -32,7 +40,10 @@ __all__ = [
     "FalFlux2ProEditGenerator",
     "FalFluxProKontextGenerator",
     "FalFluxProUltraGenerator",
+    "FalGemini25FlashImageEditGenerator",
     "FalGemini25FlashImageGenerator",
+    "FalGptImage15EditGenerator",
+    "FalGptImage15Generator",
     "FalGptImage1EditImageGenerator",
     "FalGptImage1MiniGenerator",
     "FalIdeogramCharacterGenerator",
@@ -46,4 +57,7 @@ __all__ = [
     "FalNanoBananaProEditGenerator",
     "FalQwenImageEditGenerator",
     "FalQwenImageGenerator",
+    "FalReveEditGenerator",
+    "FalReveTextToImageGenerator",
+    "FalSeedreamV45TextToImageGenerator",
 ]

package/templates/api/src/boards/generators/implementations/fal/image/bytedance_seedream_v45_edit.py

@@ -0,0 +1,219 @@
+"""
+fal.ai ByteDance Seedream v4.5 Edit image editing generator.
+
+Edit images using fal.ai's ByteDance Seedream v4.5 Edit model.
+A new-generation image creation model that integrates image generation
+and image editing capabilities into a single, unified architecture.
+Supports editing up to 10 input images with a text prompt.
+
+See: https://fal.ai/models/fal-ai/bytedance/seedream/v4.5/edit
+"""
+
+import os
+from typing import Literal
+
+from pydantic import BaseModel, Field
+
+from ....artifacts import ImageArtifact
+from ....base import BaseGenerator, GeneratorExecutionContext, GeneratorResult
+
+# Valid image size presets
+ImageSizePreset = Literal[
+    "square_hd",
+    "square",
+    "portrait_4_3",
+    "portrait_16_9",
+    "landscape_4_3",
+    "landscape_16_9",
+    "auto_2K",
+    "auto_4K",
+]
+
+
+class BytedanceSeedreamV45EditInput(BaseModel):
+    """Input schema for ByteDance Seedream v4.5 Edit.
+
+    Artifact fields (like image_sources) are automatically detected via type
+    introspection and resolved from generation IDs to ImageArtifact objects.
+    """
+
+    prompt: str = Field(description="The text prompt used to edit the image")
+    image_sources: list[ImageArtifact] = Field(
+        description="List of input images for editing (up to 10 images)",
+        min_length=1,
+        max_length=10,
+    )
+    num_images: int = Field(
+        default=1,
+        ge=1,
+        le=6,
+        description="Number of images to generate",
+    )
+    image_size: ImageSizePreset | None = Field(
+        default=None,
+        description=(
+            "The size of the generated image. Options: square_hd, square, "
+            "portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9, "
+            "auto_2K, auto_4K. Default is 2048x2048"
+        ),
+    )
+    seed: int | None = Field(
+        default=None,
+        description="Random seed to control the stochasticity of image generation",
+    )
+    sync_mode: bool = Field(
+        default=False,
+        description=(
+            "If True, the media will be returned as a data URI and the output "
+            "data won't be available in the request history"
+        ),
+    )
+    enable_safety_checker: bool = Field(
+        default=True,
+        description="Enables safety filtering on generated images",
+    )
+
+
+class FalBytedanceSeedreamV45EditGenerator(BaseGenerator):
+    """ByteDance Seedream v4.5 Edit image editing generator using fal.ai."""
+
+    name = "fal-bytedance-seedream-v45-edit"
+    artifact_type = "image"
+    description = "Fal: ByteDance Seedream v4.5 Edit - Unified image generation and editing"
+
+    def get_input_schema(self) -> type[BytedanceSeedreamV45EditInput]:
+        return BytedanceSeedreamV45EditInput
+
+    async def generate(
+        self, inputs: BytedanceSeedreamV45EditInput, context: GeneratorExecutionContext
+    ) -> GeneratorResult:
+        """Edit images using fal.ai ByteDance Seedream v4.5 Edit model."""
+        # Check for API key (fal-client uses FAL_KEY environment variable)
+        if not os.getenv("FAL_KEY"):
+            raise ValueError("API configuration invalid. Missing FAL_KEY environment variable")
+
+        # Import fal_client
+        try:
+            import fal_client
+        except ImportError as e:
+            raise ImportError(
+                "fal.ai SDK is required for FalBytedanceSeedreamV45EditGenerator. "
+                "Install with: pip install weirdfingers-boards[generators-fal]"
+            ) from e
+
+        # Upload image artifacts to Fal's public storage
+        # Fal API requires publicly accessible URLs, but our storage_url might be:
+        # - Localhost URLs (not publicly accessible)
+        # - Private S3 buckets (not publicly accessible)
+        # So we upload to Fal's temporary storage first
+        from ..utils import upload_artifacts_to_fal
+
+        image_urls = await upload_artifacts_to_fal(inputs.image_sources, context)
+
+        # Prepare arguments for fal.ai API
+        arguments: dict = {
+            "prompt": inputs.prompt,
+            "image_urls": image_urls,
+            "num_images": inputs.num_images,
+            "sync_mode": inputs.sync_mode,
+            "enable_safety_checker": inputs.enable_safety_checker,
+        }
+
+        # Add optional parameters
+        if inputs.image_size is not None:
+            arguments["image_size"] = inputs.image_size
+
+        if inputs.seed is not None:
+            arguments["seed"] = inputs.seed
+
+        # Submit async job and get handler
+        handler = await fal_client.submit_async(
+            "fal-ai/bytedance/seedream/v4.5/edit",
+            arguments=arguments,
+        )
+
+        # Store the external job ID for tracking
+        await context.set_external_job_id(handler.request_id)
+
+        # Stream progress updates (sample every 3rd event to avoid spam)
+        from .....progress.models import ProgressUpdate
+
+        event_count = 0
+        async for event in handler.iter_events(with_logs=True):
+            event_count += 1
+
+            # Process every 3rd event to provide feedback without overwhelming
+            if event_count % 3 == 0:
+                # Extract logs if available
+                logs = getattr(event, "logs", None)
+                if logs:
+                    # Join log entries into a single message
+                    if isinstance(logs, list):
+                        message = " | ".join(str(log) for log in logs if log)
+                    else:
+                        message = str(logs)
+
+                    if message:
+                        await context.publish_progress(
+                            ProgressUpdate(
+                                job_id=handler.request_id,
+                                status="processing",
+                                progress=50.0,  # Approximate mid-point progress
+                                phase="processing",
+                                message=message,
+                            )
+                        )
+
+        # Get final result
+        result = await handler.get()
+
+        # Extract image URLs from result
+        # fal.ai returns: {
+        #   "images": [{"url": "...", "width": ..., "height": ..., ...}, ...]
+        # }
+        images = result.get("images", [])
+
+        if not images:
+            raise ValueError("No images returned from fal.ai API")
+
+        # Store each image using output_index
+        artifacts = []
+        for idx, image_data in enumerate(images):
+            image_url = image_data.get("url")
+            # Extract dimensions if available, otherwise use sensible defaults
+            width = image_data.get("width", 2048)
+            height = image_data.get("height", 2048)
+
+            if not image_url:
+                raise ValueError(f"Image {idx} missing URL in fal.ai response")
+
+            # Determine format from content_type or default to png
+            content_type = image_data.get("content_type", "image/png")
+            if "jpeg" in content_type or "jpg" in content_type:
+                format_type = "jpeg"
+            elif "webp" in content_type:
+                format_type = "webp"
+            else:
+                format_type = "png"
+
+            # Store with appropriate output_index
+            artifact = await context.store_image_result(
+                storage_url=image_url,
+                format=format_type,
+                width=width,
+                height=height,
+                output_index=idx,
+            )
+            artifacts.append(artifact)
+
+        return GeneratorResult(outputs=artifacts)
+
+    async def estimate_cost(self, inputs: BytedanceSeedreamV45EditInput) -> float:
+        """Estimate cost for ByteDance Seedream v4.5 Edit generation.
+
+        Pricing not disclosed in documentation, using conservative estimate
+        based on similar high-quality image editing models.
+        """
+        # Conservative estimate per image
+        per_image_cost = 0.05
+        return per_image_cost * inputs.num_images

package/templates/api/src/boards/generators/implementations/fal/image/gemini_25_flash_image_edit.py

@@ -0,0 +1,208 @@
+"""
+Google Gemini 2.5 Flash Image edit image-to-image generator.
+
+Google's state-of-the-art image generation and editing model available through fal.ai.
+Performs image-to-image transformations and edits based on text prompts.
+Supports multiple aspect ratios and output formats with batch generation up to 4 images.
+
+Based on Fal AI's fal-ai/gemini-25-flash-image/edit model.
+See: https://fal.ai/models/fal-ai/gemini-25-flash-image/edit
+"""
+
+import os
+from typing import Literal
+
+from pydantic import BaseModel, Field
+
+from ....artifacts import ImageArtifact
+from ....base import BaseGenerator, GeneratorExecutionContext, GeneratorResult
+
+
+class Gemini25FlashImageEditInput(BaseModel):
+    """Input schema for Gemini 2.5 Flash Image edit generation.
+
+    Artifact fields (like image_sources) are automatically detected via type
+    introspection and resolved from generation IDs to ImageArtifact objects.
+    """
+
+    prompt: str = Field(
+        description="The editing instruction for image transformation",
+        min_length=3,
+        max_length=5000,
+    )
+    image_sources: list[ImageArtifact] = Field(
+        description="List of input images for editing (from previous generations)",
+        min_length=1,
+    )
+    num_images: int = Field(
+        default=1,
+        ge=1,
+        le=4,
+        description="Number of images to generate (max 4)",
+    )
+    aspect_ratio: (
+        Literal[
+            "auto",
+            "21:9",
+            "16:9",
+            "3:2",
+            "4:3",
+            "5:4",
+            "1:1",
+            "4:5",
+            "3:4",
+            "2:3",
+            "9:16",
+        ]
+        | None
+    ) = Field(
+        default="auto",
+        description="Image aspect ratio. Default 'auto' uses input image's aspect ratio.",
+    )
+    output_format: Literal["jpeg", "png", "webp"] = Field(
+        default="png",
+        description="Output image format",
+    )
+    sync_mode: bool = Field(
+        default=False,
+        description="Return media as data URI without request history storage",
+    )
+    limit_generations: bool = Field(
+        default=False,
+        description="Restrict to single generation per round (experimental)",
+    )
+
+
+class FalGemini25FlashImageEditGenerator(BaseGenerator):
+    """Google Gemini 2.5 Flash Image edit generator using fal.ai."""
+
+    name = "fal-gemini-25-flash-image-edit"
+    artifact_type = "image"
+    description = "Fal: Gemini 2.5 Flash Image Edit - AI-powered image editing with Gemini"
+
+    def get_input_schema(self) -> type[Gemini25FlashImageEditInput]:
+        return Gemini25FlashImageEditInput
+
+    async def generate(
+        self, inputs: Gemini25FlashImageEditInput, context: GeneratorExecutionContext
+    ) -> GeneratorResult:
+        """Edit images using Google Gemini 2.5 Flash Image via fal.ai."""
+        # Check for API key (fal-client uses FAL_KEY environment variable)
+        if not os.getenv("FAL_KEY"):
+            raise ValueError("API configuration invalid. Missing FAL_KEY environment variable")
+
+        # Import fal_client
+        try:
+            import fal_client
+        except ImportError as e:
+            raise ImportError(
+                "fal.ai SDK is required for FalGemini25FlashImageEditGenerator. "
+                "Install with: pip install weirdfingers-boards[generators-fal]"
+            ) from e
+
+        # Upload image artifacts to Fal's public storage
+        # Fal API requires publicly accessible URLs, but our storage_url might be:
+        # - Localhost URLs (not publicly accessible)
+        # - Private S3 buckets (not publicly accessible)
+        # So we upload to Fal's temporary storage first
+        from ..utils import upload_artifacts_to_fal
+
+        image_urls = await upload_artifacts_to_fal(inputs.image_sources, context)
+
+        # Prepare arguments for fal.ai API
+        arguments = {
+            "prompt": inputs.prompt,
+            "image_urls": image_urls,
+            "num_images": inputs.num_images,
+            "output_format": inputs.output_format,
+            "sync_mode": inputs.sync_mode,
+            "limit_generations": inputs.limit_generations,
+        }
+
+        # Add aspect_ratio if provided
+        if inputs.aspect_ratio is not None:
+            arguments["aspect_ratio"] = inputs.aspect_ratio
+
+        # Submit async job and get handler
+        handler = await fal_client.submit_async(
+            "fal-ai/gemini-25-flash-image/edit",
+            arguments=arguments,
+        )
+
+        # Store the external job ID for tracking
+        await context.set_external_job_id(handler.request_id)
+
+        # Stream progress updates (sample every 3rd event to avoid spam)
+        from .....progress.models import ProgressUpdate
+
+        event_count = 0
+        async for event in handler.iter_events(with_logs=True):
+            event_count += 1
+
+            # Process every 3rd event to provide feedback without overwhelming
+            if event_count % 3 == 0:
+                # Extract logs if available
+                logs = getattr(event, "logs", None)
+                if logs:
+                    # Join log entries into a single message
+                    if isinstance(logs, list):
+                        message = " | ".join(str(log) for log in logs if log)
+                    else:
+                        message = str(logs)
+
+                    if message:
+                        await context.publish_progress(
+                            ProgressUpdate(
+                                job_id=handler.request_id,
+                                status="processing",
+                                progress=50.0,  # Approximate mid-point progress
+                                phase="processing",
+                                message=message,
+                            )
+                        )
+
+        # Get final result
+        result = await handler.get()
+
+        # Extract image URLs and description from result
+        # fal.ai returns: {
+        #   "images": [{"url": "...", "width": ..., "height": ..., ...}, ...],
+        #   "description": "Text description from Gemini"
+        # }
+        images = result.get("images", [])
+
+        if not images:
+            raise ValueError("No images returned from fal.ai API")
+
+        # Store each image using output_index
+        artifacts = []
+        for idx, image_data in enumerate(images):
+            image_url = image_data.get("url")
+            # Use 'or' to handle explicit None values from API
+            width = image_data.get("width") or 1024
+            height = image_data.get("height") or 1024
+
+            if not image_url:
+                raise ValueError(f"Image {idx} missing URL in fal.ai response")
+
+            # Store with appropriate output_index
+            artifact = await context.store_image_result(
+                storage_url=image_url,
+                format=inputs.output_format,
+                width=width,
+                height=height,
+                output_index=idx,
+            )
+            artifacts.append(artifact)
+
+        return GeneratorResult(outputs=artifacts)
+
+    async def estimate_cost(self, inputs: Gemini25FlashImageEditInput) -> float:
+        """Estimate cost for Gemini 2.5 Flash Image edit generation.
+
+        Note: Pricing information not available in fal.ai documentation.
+        Using placeholder estimate similar to other Gemini-based models.
+        """
+        # Placeholder cost estimate per image (similar to nano-banana which also uses Gemini)
+        per_image_cost = 0.039
+        return per_image_cost * inputs.num_images