Alibaba's Wan 2.7 is a production-ready suite of AI image-generation and image-editing endpoints, featuring a built-in thinking mode for enhanced reasoning and higher-quality outputs. Wan 2.7 covers two core workflows: text-to-image generation and image editing, each available in standard and pro tiers, offering flexible quality-cost trade-offs.
Wan 2.7 Series — Text-to-Image & Image Editing API
Wan 2.7 offers four focused endpoints for generating images from text or editing existing images—ideal for commercial pipelines, creative production, and repeatable content workflows.
- Wan 2.7 Text-to-Image — Generate high-quality images from text prompts using built-in thinking mode for smarter composition and accurate prompt adherence.
- Wan 2.7 Image Edit — Edit and refine images with precise, controlled modifications while preserving structure and subject consistency.
- Wan 2.7 Text-to-Image Pro — Pro tier with up to 4K (4096×4096) output for print-ready and large-format assets.
- Wan 2.7 Image Edit Pro — Pro tier image editing with up to 2K resolution for high-fidelity results.
Key Features
- Thinking Mode — Built-in chain-of-thought reasoning delivers better prompt adherence, more coherent compositions, and fewer artifacts.
- Up to 4K Image Output — Pro tier supports up to 4096×4096 resolution, suitable for print and large-format display.
- Multi-Image Reference Editing — Upload 1–9 reference images for style, subject, and background guidance in editing workflows.
- Standard & Pro Tiers — Choose cost-efficient standard or higher-quality pro endpoints based on your use case.
- Seed-Based Reproducibility — Flexible sizing and seed control for fast iteration and repeatable creative testing.