Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation, is now available on SiliconFlow.
Flexible Output Resolutions: I2V supports 720P video output; T2V supports 480P & 720P resolutions
SiliconFlow-exclusive Turbo versions: Reduce 30% video generation time, ideal for high-throughput applications
Budget-friendly pricing:
Wan2.1-I2V-14B-720P & Wan2.1-T2V-14B ($0.29 / video)
Wan2.1-I2V-14B-720P-Turbo & Wan2.1-T2V-14B-Turbo ($0.21 / video)
Key Capabilities
Leveraging the advanced Diffusion Transformer (DiT) architecture, Spatio-Temporal VAE, and large-scale scalable pretraining, Wan2.1-I2V-14B and Wan2.1-T2V-14B excel at generating high-quality, dynamic videos.
Wan2.1-I2V-14B
Multi-Modal Input Processing:Generates videos from both text prompts and image inputs
Complex Visual Scene Creation:Excels at translating static images into dynamic content with intricate motion patterns
Wan2.1-T2V-14B
Bilingual Language Support - Unique capability to process both Chinese and English text inputs
Superior Motion Dynamics - Produces exceptional visuals with substantial and realistic motion complexity
Turbo VersionsTea
Cache Technology: Training-free caching method that accelerates inference by leveraging timestep fluctuation differences
30% Speed Boost: Faster video generation while maintaining quality for high-throughput applications
Cost-Effective: Enhanced performance at lower pricing ($0.21 vs $0.29 per video)
Quick Start
Try the Wan2.1 series on the SiliconFlow playground.
Quick Access to API
The following Python example illustrates how to invoke SiliconFlow’s API to generate video using the Wan2.1-I2V-14B-720P (Turbo) model.
Whether you're crafting cinematic content, building interactive experiences, or developing next-generation media application, Wan2.1 empowers developers and creators to bring any concept to life with unprecedented quality and precision.
Ready to revolutionize your video generation workflows? Experience Wan2.1 today on SiliconFlow.