--- language: - en license: apache-2.0 pipeline_tag: image-to-video library_name: ovi base_model: - Wan-AI/Wan2.2-TI2V-5B ---

Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation

[Chetwin Low](https://www.linkedin.com/in/chetwin-low-061975193/) * 1 , [Weimin Wang](https://www.linkedin.com/in/weimin-wang-will/) * † 1 , [Calder Katyal](https://www.linkedin.com/in/calder-katyal-a8a9b3225/) 2
* Equal contribution, Project Lead
1 Character AI, 2 Yale University
--- ## 🎥 Video Demo ### 🆕 Ovi 1.1 10-Second Demo

Ovi 1.1 – 10-second temporally consistent video generation (960 × 960 resolution)

### 🎬 Original 5-Second Demo
--- # 🆕 Ovi 1.1 Update (10 November 2025) - **Key Feature:** Enables *temporal-consistent 10-second video generation* at **960 × 960 resolution** - **Training Improvements:** - Trained natively on 960×960 resolution videos - Dataset includes **100% more videos** for greater diversity - - **Prompt Format Update:** - Audio descriptions should now be written as ``` Audio: ... ``` instead of using ``` ... ``` ## 🌟 Key Features Ovi is a veo-3-like, **video + audio generation model** that simultaneously generates both video and audio content from text or text + image inputs. - **🎬 Video+Audio Generation**: Generate synchronized video and audio content simultaneously - **🎵 High-Quality Audio Branch**: We designed and pretrained our 5B audio branch from scratch using our high quality in-house audio datasets - **📝 Flexible Input**: Supports text-only or text+image conditioning - **⏱️ 10-second (or 5-second) Videos**: Generates 10-second or 5-second videos at 24 FPS, resolution of 960x960p, at various aspect ratios (9:16, 16:9, 1:1, etc) - **🔧 ComfyUI Integration**: ComfyUI support is now available via [ComfyUI-WanVideoWrapper](https://github.com/kijai/ComfyUI-WanVideoWrapper), related [PR](https://github.com/kijai/ComfyUI-WanVideoWrapper/issues/1343#issuecomment-3382969479). - **🎬 Create videos now on wavespeed.ai**: https://wavespeed.ai/models/character-ai/ovi/image-to-video & https://wavespeed.ai/models/character-ai/ovi/text-to-video - **🎬 Create videos now on HuggingFace**: https://huggingface.co/spaces/akhaliq/Ovi ### 🎯 10-second examples

Click the ⛶ button on any video to view full screen.

### 🎯 5-second examples

Click the ⛶ button on any video to view full screen.

--- ## 📋 Todo List - [x] Release research paper and [website for demos](https://aaxwaz.github.io/Ovi) - [x] Checkpoint of 11B model - [x] Inference Codes - [x] Text or Text+Image as input - [x] Gradio application code - [x] Multi-GPU inference with or without the support of sequence parallel - [x] fp8 weights and improved memory efficiency (credits to [@rkfg](https://github.com/rkfg)) - [x] qint8 quantization thanks to [@gluttony-10](https://github.com/character-ai/Ovi/commits?author=gluttony-10) - [ ] Improve efficiency of Sequence Parallel implementation - [ ] Implement Sharded inference with FSDP - [x] Video creation example prompts and format - [x] Finetune model with higher resolution data, and RL for performance improvement. - [x] Longer video generation (10s) - [ ] Reference voice condition - [ ] Distilled model for faster inference - [ ] Training scripts --- ## 🎨 An Easy Way to Create We provide example prompts to help you get started with Ovi: - **Text-to-Audio-Video (T2AV) 10s**: [`example_prompts/gpt_examples_t2v.csv`](example_prompts/gpt_examples_10s_t2v.csv) - **Image-to-Audio-Video (I2AV) 10s**: [`example_prompts/gpt_examples_i2v.csv`](example_prompts/gpt_examples_10s_i2v.csv) - **Text-to-Audio-Video (T2AV)**: [`example_prompts/gpt_examples_t2v.csv`](example_prompts/gpt_examples_t2v.csv) - **Image-to-Audio-Video (I2AV)**: [`example_prompts/gpt_examples_i2v.csv`](example_prompts/gpt_examples_i2v.csv) ### 📝 Prompt Format Our prompts use special tags to control speech and audio: - **Speech**: `Your speech content here` - Text enclosed in these tags will be converted to speech - **Audio Description**: `Audio: YOUR AUDIO DESCRIPTION` - Describes the audio or sound effects present in the video **at the end of prompt!** --- ## 📦 Installation ### Step-by-Step Installation ```bash # Clone the repository git clone https://github.com/character-ai/Ovi.git cd Ovi # Create and activate virtual environment virtualenv ovi-env source ovi-env/bin/activate # Install PyTorch first pip install torch==2.6.0 torchvision torchaudio # Install other dependencies pip install -r requirements.txt # Install Flash Attention pip install flash_attn --no-build-isolation ``` ### Alternative Flash Attention Installation (Optional) If the above flash_attn installation fails, you can try the Flash Attention 3 method: ```bash git clone https://github.com/Dao-AILab/flash-attention.git cd flash-attention/hopper python setup.py install cd ../.. # Return to Ovi directory ``` ## Download Weights To download our main Ovi checkpoint, as well as T5 and vae decoder from Wan, and audio vae from MMAudio ``` # Default is downloaded to ./ckpts, and the inference yaml is set to ./ckpts so no change required # Default installs all versions of Ovi models, 720x720_5s, 960x960_5s, 960x960_10s python3 download_weights.py # For qint8 also ues python3 download_weights.py OR # Optional can specific --output-dir to download to a specific directory # but if a custom directory is used, the inference yaml has to be updated with the custom directory python3 download_weights.py --output-dir # Optional can specific --models to download selective versions of Ovi instead of all of them # but if a custom directory is used, the inference yaml has to be updated with the custom directory python3 download_weights.py --models 960x960_10s # ["720x720_5s", "960x960_5s", "960x960_10s"] # Additionally, if you only have ~ 24Gb of GPU vram, please download the fp8 quantized version of the model, and follow the following instructions in sections below to run with fp8 wget -O "./ckpts/Ovi/model_fp8_e4m3fn.safetensors" "https://huggingface.co/rkfg/Ovi-fp8_quantized/resolve/main/model_fp8_e4m3fn.safetensors" ``` ## 🚀 Run Examples ### ⚙️ Configure Ovi Ovi's behavior and output can be customized by modifying [ovi/configs/inference/inference_fusion.yaml](ovi/configs/inference/inference_fusion.yaml) configuration file. The following parameters control generation quality, video resolution, and how text, image, and audio inputs are balanced: ```yaml # Output and Model Configuration model_name: "960x960_10s" # ["720x720_5s", "960x960_5s", "960x960_10s"] output_dir: "/path/to/save/your/videos" # Directory to save generated videos ckpt_dir: "/path/to/your/ckpts/dir" # Path to model checkpoints # Generation Quality Settings sample_steps: 50 # Number of denoising steps. Lower (30-40) = faster generation solver_name: "unipc" # Sampling algorithm for denoising process shift: 5.0 # Timestep shift factor for sampling scheduler seed: 100 # Random seed for reproducible results # Guidance Strength Control audio_guidance_scale: 3.0 # Strength of audio conditioning. Higher = better audio-text sync video_guidance_scale: 4.0 # Strength of video conditioning. Higher = better video-text adherence slg_layer: 11 # Layer for applying SLG (Skip Layer Guidance) technique - feel free to try different layers! # Multi-GPU and Performance sp_size: 1 # Sequence parallelism size. Set equal to number of GPUs used cpu_offload: False # CPU offload, will largely reduce peak GPU VRAM but increase end to end runtime by ~20 seconds fp8: False # load fp8 version of model, will have quality degradation and will not have speed up in inference time as it still uses bf16 matmuls, but can be paired with cpu_offload=True, to run model with 24Gb of GPU vram # Input Configuration text_prompt: "/path/to/csv" or "your prompt here" # Text prompt OR path to CSV/TSV file with prompts mode: ['i2v', 't2v', 't2i2v'] # Generate t2v, i2v or t2i2v; if t2i2v, it will use flux krea to generate starting image and then will follow with i2v video_frame_height_width: [704, 1280] # Video dimensions [height, width] for T2V mode only each_example_n_times: 1 # Number of times to generate each prompt # Quality Control (Negative Prompts) video_negative_prompt: "jitter, bad hands, blur, distortion" # Artifacts to avoid in video audio_negative_prompt: "robotic, muffled, echo, distorted" # Artifacts to avoid in audio ``` ### 🎬 Running Inference #### **Single GPU** (Simple Setup) ```bash python3 inference.py --config-file ovi/configs/inference/inference_fusion.yaml ``` *Use this for single GPU setups. The `text_prompt` can be a single string or path to a CSV file.* #### **Multi-GPU** (Parallel Processing) ```bash torchrun --nnodes 1 --nproc_per_node 8 inference.py --config-file ovi/configs/inference/inference_fusion.yaml ``` *Use this to run samples in parallel across multiple GPUs for faster processing.* ### Memory & Performance Requirements Below are approximate GPU memory requirements for different configurations. Sequence parallel implementation will be optimized in the future. All End-to-End time calculated based on a 121 frame, 720x720 video, using 50 denoising steps. Minimum GPU vram requirement to run our model is **32Gb**, fp8 parameters is currently supported, reducing peak VRAM usage to **24Gb** with slight quality degradation. | Sequence Parallel Size | FlashAttention-3 Enabled | CPU Offload | With Image Gen Model | Peak VRAM Required | End-to-End Time | |-------------------------|---------------------------|-------------|-----------------------|---------------|-----------------| | 1 | Yes | No | No | ~80 GB | ~83s | | 1 | No | No | No | ~80 GB | ~96s | | 1 | Yes | Yes | No | ~80 GB | ~105s | | 1 | No | Yes | No | ~32 GB | ~118s | | **1** | **Yes** | **Yes** | **Yes** | **~32 GB** | **~140s** | | 4 | Yes | No | No | ~80 GB | ~55s | | 8 | Yes | No | No | ~80 GB | ~40s | ### Gradio We provide a simple script to run our model in a gradio UI. It uses the `ckpt_dir` in `ovi/configs/inference/inference_fusion.yaml` to initialize the model ```bash python3 gradio_app.py OR # To enable cpu offload to save GPU VRAM, will slow down end to end inference by ~20 seconds python3 gradio_app.py --cpu_offload OR # To enable an additional image generation model to generate first frames for I2V, cpu_offload is automatically enabled if image generation model is enabled python3 gradio_app.py --use_image_gen OR # To run model with 24Gb GPU vram. No need to download additional models. python3 gradio_app.py --cpu_offload --qint8 # To run model with 24Gb GPU vram python3 gradio_app.py --cpu_offload --fp8 ``` --- ## 🙏 Acknowledgements We would like to thank the following projects: - **[Wan2.2](https://github.com/Wan-Video/Wan2.2)**: Our video branch is initialized from the Wan2.2 repository - **[MMAudio](https://github.com/hkchengrex/MMAudio)**: We reused MMAudio's audio vae. --- ## 🤝 Collaboration We welcome all types of collaboration! Whether you have feedback, want to contribute, or have any questions, please feel free to reach out. **Contact**: [Weimin Wang](https://linkedin.com/in/weimin-wang-will) for any issues or feedback. ## ⭐ Citation If Ovi is helpful, please help to ⭐ the repo. If you find this project useful for your research, please consider citing our [paper](https://arxiv.org/abs/2510.01284). ### BibTeX ```bibtex @misc{low2025ovitwinbackbonecrossmodal, title={Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation}, author={Chetwin Low and Weimin Wang and Calder Katyal}, year={2025}, eprint={2510.01284}, archivePrefix={arXiv}, primaryClass={cs.MM}, url={https://arxiv.org/abs/2510.01284}, } ```