Modelpublic

wavespeed/Ovi-e4m3_e4m3_dynamic_per_tensor

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

1likes2downloads

Model Card

About this repository

FP8 (e4m3) quantized Ovi fusion model, published by WaveSpeed AI. Ovi generates video and its matching audio in one pass; both towers are quantized here — 30 video blocks and 30 audio blocks — with e4m3_e4m3_dynamic_per_tensor on attention and FFN layers and e4m3_weightonly elsewhere. Compute stays in bf16. FP8 matmul needs Hopper (H100/H200) or newer to be faster than bf16.

The original model is Ovi by Character.AI, released under Apache-2.0; that license carries over to these weights.

Ovi FusionModel - FP8 Quantized

This is the Ovi FusionModel quantized with FP8 (e4m3e4m3dynamicpertensor) for faster inference.

Quantization Details

—Video Model Blocks: 30 blocks quantized
—Audio Model Blocks: 30 blocks quantized
—Attention/FFN layers: e4m3e4m3dynamicpertensor
—Other layers: e4m3_weightonly

Usage

python

import sys
import os
import torch
from omegaconf import OmegaConf
from huggingface_hub import hf_hub_download

OVI_PATH = "./workspace/Ovi"
sys.path.insert(0, OVI_PATH)
os.chdir(OVI_PATH)

from ovi.ovi_fusion_engine import OviFusionEngine

# Download quantized weights
model_path = hf_hub_download(
    repo_id="wavespeed/Ovi-e4m3_e4m3_dynamic_per_tensor",
    filename="model.pth"
)

config = OmegaConf.load("config.yaml")
engine = OviFusionEngine(config=config, device="cuda", target_dtype=torch.bfloat16)

# Load quantized weights
engine.model.load_state_dict(torch.load(model_path))

# Model is already quantized, ready for inference

Model Card

—Developed by: Alibaba/Character.AI
—Model type: Video + Audio generation (FusionModel)
—Quantization: FP8 (e4m3e4m3dynamicpertensor)
—License: Check original Ovi repository

Original Model

Based on Ovi