Hybrid 720P text- and image-to-video
According to the model card, TI2V-5B supports both text-to-video and image-to-video generation at 720P resolution with 24fps.
Open Source Model Profile · Wan-AI
Wan2.2-TI2V-5B
Wan2.2-TI2V-5B is a 5B Wan-AI model for text-to-video and image-to-video generation. According to the model card, it produces 720P video at 24fps and can run on a single consumer-grade GPU.
- Publisher
- Wan-AI
- Task
- text-to-video
- Model type
- ti2v
- License
- apache-2.0
- Library
- wan2.2
- Publication status
- Accepted · not indexed
Model overview
Wan2.2-TI2V-5B is published by Wan-AI as a text-to-video model. The captured record lists model type ti2v with the wan2.2 library and an apache-2.0 license. According to the model card, this repository contains the 5B TI2V model for hybrid text-to-video and image-to-video generation at 720P and 24fps.
Recorded capabilities
High-compression VAE design
According to the model card, the model uses Wan2.2-VAE with 16x16x4 compression, extended by patchification, for efficient high-definition generation.
Documented single-GPU commands
According to the model card, single-GPU text-to-video and image-to-video commands use task ti2v-5B at 1280x704 with offload, dtype conversion, and CPU text-encoder options.
Wan2.2 generation context
According to the model card, Wan2.2 builds on Wan2.1 with MoE architecture, larger training data, and high-compression generation; the A14B MoE context is separate from this 5B dense model.
Use cases in the source record
- Text-to-video generation at 720P and 24fps from prompts using the documented ti2v-5B single-GPU or multi-GPU commands.
- Image-to-video generation by supplying an input image with a prompt, following the model card's image-parameter workflow.
Limitations and unknowns
- No Safetensors parameter count was extracted from this record beyond the publisher-stated 5B scale.
- No independent evaluation results were extracted; performance descriptions are publisher claims.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: Wan-AI/Wan2.2-TI2V-5B
Captured: Unknown. Processed: 2026-09-07T19:34:38.583187+00:00.
Wan2.2 💜 Wan ｜ 🖥️ GitHub | 🤗 Hugging Face | 🤖 ModelScope | 📑 Technical Report | 📑 Blog | 💬 WeChat Group | 📖 Discord Wan: Open and Advanced Large-Scale Video Generative Models We are excited to introduce Wan2.2 , a major upgrade to our foundational video models. With Wan2.2 , we have focused on incorporating the following innovations: 👍 Effective MoE Architecture : Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational co…
F001F002F003F004F005F006F008F009F012F013F016F017F019F023F024F026