Banana Image MCP

"""Gemini 3 Pro Image specialized service for high-quality generation.""" from typing import Any from ..config.settings import MediaResolution, ProImageConfig, ThinkingLevel from .base_image_service import BaseImageService from .gemini_client import GeminiClient from .image_storage_service import ImageStorageService class ProImageService(BaseImageService): """Service for high-quality image generation using Gemini 3 Pro Image model. Features: - Up to 4K resolution support - Google Search grounding for factual accuracy - Advanced reasoning with configurable thinking levels - Professional-grade outputs """ def __init__( self, gemini_client: GeminiClient, config: ProImageConfig, storage_service: ImageStorageService | None = None, ): """Initialize Pro image service. Args: gemini_client: Gemini API client configured for Pro model config: Pro model configuration storage_service: Optional storage service for thumbnails """ super().__init__(gemini_client, config, storage_service) self.pro_config = config def _get_operation_name(self) -> str: """Get operation name for progress tracking.""" return "pro_image_generation" def _build_generation_config( self, thinking_level: ThinkingLevel | None = None, media_resolution: MediaResolution | None = None, **kwargs: Any, ) -> dict[str, Any]: """Build generation config for Pro model. Args: thinking_level: Reasoning depth (LOW or HIGH) media_resolution: Vision processing detail level **kwargs: Additional parameters (ignored) Returns: Configuration dictionary with Pro-specific parameters """ # Apply defaults from config level = thinking_level or self.pro_config.default_thinking_level resolution = media_resolution or self.pro_config.default_media_resolution config: dict[str, Any] = { "thinking_level": level.value, } # Add media resolution if supported if self.pro_config.supports_media_resolution: config["media_resolution"] = resolution.value return config def _build_metadata( self, prompt: str, response_index: int, image_index: int, resolution: str = "high", thinking_level: ThinkingLevel | None = None, media_resolution: MediaResolution | None = None, enable_grounding: bool | None = None, negative_prompt: str | None = None, system_instruction: str | None = None, aspect_ratio: str | None = None, **kwargs: Any, ) -> dict[str, Any]: """Build metadata for Pro-generated image. Args: prompt: Generation prompt response_index: Index of the response (1-based) image_index: Index of the image within response (1-based) resolution: Output resolution setting thinking_level: Reasoning depth used media_resolution: Vision processing detail level used enable_grounding: Whether grounding was enabled negative_prompt: Optional negative prompt used system_instruction: Optional system instruction used aspect_ratio: Optional aspect ratio used **kwargs: Additional parameters (ignored) Returns: Metadata dictionary with Pro-specific fields """ # Apply defaults level = thinking_level or self.pro_config.default_thinking_level media_res = media_resolution or self.pro_config.default_media_resolution grounding = ( enable_grounding if enable_grounding is not None else self.pro_config.enable_search_grounding ) return { "model": self.pro_config.model_name, "model_tier": "pro", "response_index": response_index, "image_index": image_index, "resolution": resolution, "thinking_level": level.value, "media_resolution": media_res.value, "grounding_enabled": grounding, "prompt": prompt, "negative_prompt": negative_prompt, "system_instruction": system_instruction, "aspect_ratio": aspect_ratio, "mime_type": f"image/{self.config.default_image_format}", "synthid_watermark": True, } def _enhance_prompt( self, prompt: str, resolution: str = "high", negative_prompt: str | None = None, **kwargs: Any, ) -> str: """Enhance prompt to leverage Pro model capabilities. Pro model benefits from: - Narrative, descriptive prompts - Specific composition/lighting details - Quality and detail emphasis Args: prompt: Original prompt resolution: Target resolution negative_prompt: Optional constraints (handled separately in base class) **kwargs: Additional parameters (ignored) Returns: Enhanced prompt string """ enhanced = prompt # Pro model benefits from narrative prompts for short inputs if len(prompt) < 50: enhanced = ( f"Create a high-quality, detailed image: {prompt}. " "Pay attention to composition, lighting, and fine details." ) # Resolution hints for high-res outputs if resolution in ["4k", "high", "2k"]: if "text" in prompt.lower() or "diagram" in prompt.lower(): enhanced += " Ensure text is sharp and clearly readable at high resolution." if resolution == "4k": enhanced += " Render at maximum 4K quality with exceptional detail." return enhanced