reve/reve-create

Parameters

Pricing

Unit: $/img

Related Models

• reve/reve-edit: [Core Function] Reve Edit is Reve’s instruction-based single-image editing model. [Strengths] It excels at precise, localized edits that follow a natural-language instruction while preserving the untouched parts of the source image. [Best For] Highly recommended for: adding, removing or replacing objects, changing colors or attributes, adjusting style, and editing backgrounds on a single source image. [Limitations] Do NOT use this model to generate an image from scratch; use Reve Create for that. Do NOT use it to blend several images together; use Reve Remix for that. It requires exactly one reference image. [Routing] Choose this model when the user provides one image and wants targeted edits described in words.
• reve/reve-remix: [Core Function] Reve Remix is Reve’s multi-image composition model that blends 1 to 6 reference images guided by a text prompt. [Strengths] It excels at combining subjects, styles and elements from multiple references into one coherent image, and supports inline markers in the prompt to address specific input images by index. [Best For] Highly recommended for: merging subjects from different photos, transferring style from reference images, compositing a product into a new scene, and keeping a character consistent across references. [Limitations] Do NOT use this model for simple single-image edits; use Reve Edit for that. Do NOT use it for pure text-to-image generation; use Reve Create for that. It accepts between 1 and 6 reference images. [Routing] Choose this model when the user supplies multiple reference images to blend into a single result.
• openai/gpt-image-1.5: [Core Function] GPT Image 1.5 is a versatile text-to-image generation model. [Strengths] It balances solid visual performance with crucial utility features, notably its native support for generating images with transparent backgrounds. [Best For] Highly recommended for: creating UI icons, standalone logos, game assets, and any graphic design elements that require a transparent background. [Limitations] Do NOT use this model if you need 2K or 4K resolution. Its maximum supported resolution is 1536x1024. [Routing] Choose this model specifically when the user asks for ‘transparent background’, ‘no background’, or ‘PNG icon’. For standard, high-fidelity, or 4K image generation, use GPT Image 2 instead.
• alibaba/wan2.7-image-pro: [Core Function] Wan 2.7 Image Pro is Alibaba’s flagship reasoning-enhanced image generation model. [Strengths] It features built-in chain-of-thought reasoning (Thinking Mode), exceptional prompt accuracy, native 12-language text rendering, and generates ultra-high-resolution 4K images. [Best For] Highly recommended for: print-ready large-format posters, complex logical prompts, and generating images containing specific text/typography. [Limitations] Do NOT use this model if you need to generate batch images rapidly (use Wan 2.7 Image instead) or if you specifically need negative prompts (use Qwen Image 2.0 Pro). [Routing] Use this model by default for high-end, 4K, or text-heavy image generation tasks.
• google/imagen-4.0-generate-001: [Core Function] Imagen 4.0 is Google’s flagship professional-grade text-to-image model designed for high-fidelity visual generation. [Strengths] It excels at industry-leading photorealism, exceptional typography/text rendering inside images, and strong prompt adherence, supporting up to 2K resolution and batch generation. [Best For] Highly recommended for general high-quality image generation, marketing assets (posters, product labels, menus, signage), and product photography mockups where readable text is required. [Limitations] Do NOT use this model if you need instant/real-time generation, native image editing (inpainting/outpainting), or negative prompts. It is strictly an image-only output model. [Routing] Use this model by default for high-quality, photorealistic text-to-image requests. If the user emphasizes speed or needs lower cost, route to Imagen 4.0 Fast. If the user demands the absolute highest detail and prompt precision, route to Imagen 4.0 Ultra.
• alibaba/qwen-image-2.0-pro: [Core Function] Qwen Image 2.0 Pro is a professional-grade, highly controllable image generation model. [Strengths] It provides exquisite photorealism, professional infographics generation, supports NEGATIVE prompts, and can generate up to 6 image variants per API call. [Best For] Highly recommended for: workflows requiring strict negative prompt exclusion, batch generation (6 variants), and highly detailed infographics. [Limitations] Does not natively output 4K resolution (max is 2K). Does not have the explicit ‘Thinking Mode’ of Wan 2.7. [Routing] Route to this model specifically when the user provides a ‘negative prompt’ or asks for ‘batch generation of 6 images’.
• bytedance/seedream-5.0-lite: [Core Function] Seedream 5.0 Lite is a smart, reasoning-enhanced image generation model with real-time web search capabilities. [Strengths] It excels at generating time-sensitive imagery, infographics, and content requiring deep world knowledge or online search, boasting superior prompt understanding and reasoning. [Best For] Highly recommended for: current-event posters, text-heavy designs, and concept art requiring complex logical reasoning. [Limitations] As a ‘Lite’ model, its absolute photorealistic aesthetic ceiling might be slightly lower than the specialized 4.5 model. [Routing] Use this model by default when the user needs real-time information, deep reasoning, or complex intent understanding in their image.
• google/imagen-4.0-ultra-generate-001: [Core Function] Imagen 4.0 Ultra is Google’s premium text-to-image model optimized for the highest quality and instruction alignment. [Strengths] It delivers ultimate photorealistic detail, superior texture rendering, and extremely strict adherence to complex, multi-part prompt instructions, supporting up to 2K resolution. [Best For] Highly recommended for high-end commercial marketing assets, professional design, intricate artistic compositions, and scenarios where precise instruction-following is paramount. [Limitations] Do NOT use this model if you need instant/real-time generation, native image editing (inpainting/outpainting), or negative prompts. It is the most expensive tier in the Imagen family. [Routing] Choose this model when the user demands the absolute highest quality, ultimate detail, or has highly complex prompt instructions. Otherwise, default to standard Imagen 4.0 or Imagen 4.0 Fast.