Skip to content

腾讯推出的多模态驱动的定制化视频生成框架,让视频创作更自由 支持文本、图像、音频、视频等多种输入模态,满足多样化创作需求;单主体视频定制效果出色,身份一致性高达0.627;提供单GPU和多GPU推理方案,适配不同硬件环境

active 2025-05-082025-05-21 (UTC)

Partial coverage11,672 / 12,880 hourly files (91%) · 2 absent upstream · 1,204 failed, retryable2025-02-202026-08-10 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
536
Pushes
5
Pull requests
1
Issues
12
Stars
465
Forks
35

Activity over time

Daily event counts in the loaded window

Line chart, 14 days from 2025-05-08 to 2025-05-21. Pushes: 5 total, peak 3 in a day. Pull requests: 1 total, peak 1 in a day. Issues: 12 total, peak 3 in a day. Comments: 17 total, peak 4 in a day. Stars: 465 total, peak 160 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

Recent activity

Latest issues, pull requests and releases

  • Issue#20rohitpaul242025-05-20 04:03
    Output single image
  • Issue#19Marcophono22025-05-18 20:32
    8x RTX 4090 -> OOM
  • Issue comment#18gessyoo2025-05-18 18:46
    Output just says "Killed" with no other information
  • Issue comment#5CazCzworkowski-90North2025-05-16 00:36
    Has the code for audio generation been open-sourced yet?
  • Issue comment#8Cazforshort2025-05-16 00:07
    OOM Error on Windows 11 24GB GPU (RTX 3090) with sample_gpu_poor.py Despite Following Low VRAM Guidelines
  • Issue#17WYCAS2025-05-15 03:48
    请问最大支持的输入为多大
  • Pull request#16zhouzhengguang2025-05-15 02:52
  • Issue comment#10zhouzhengguang2025-05-15 02:31
    Comfyui workflow
  • Issue#15Hickey82025-05-14 05:14
    Infer without cpu_offload
  • Issue comment#13zhouzhengguang2025-05-14 02:22
    video driven edit confused
  • Issue comment#13zhouzhengguang2025-05-13 03:57
    video driven edit confused
  • Issue comment#10zhouzhengguang2025-05-13 03:49
    Comfyui workflow
  • Issue comment#4zhouzhengguang2025-05-13 03:14
    only support one image as reference?
  • Issue comment#4lsh2-meitu2025-05-13 02:36
    only support one image as reference?
  • Issue comment#6zhouzhengguang2025-05-12 01:58
    运行低显存代码时 出现错误
  • Issue comment#4zhouzhengguang2025-05-12 01:56
    only support one image as reference?
  • Issue#11AndrewBeniston2025-05-11 20:44
    Performance Discrepancy and Guidance for Single A100 GPU Gradio Setup (HunyuanCustom)
  • Issue comment#9petermg2025-05-11 15:20
    video_buffer = ret_dict['content'][0]['buffer']
  • Issue#10loboere2025-05-11 01:02
    Comfyui workflow
  • Issue comment#8AndrewBeniston2025-05-11 00:31
    OOM Error on Windows 11 24GB GPU (RTX 3090) with sample_gpu_poor.py Despite Following Low VRAM Guidelines
  • Issue#9nxneo2025-05-11 00:01
    video_buffer = ret_dict['content'][0]['buffer']
  • Issue comment#6FBAdmirer2025-05-10 21:29
    运行低显存代码时 出现错误
  • Issue#8Vibecoder90002025-05-10 19:38
    OOM Error on Windows 11 24GB GPU (RTX 3090) with sample_gpu_poor.py Despite Following Low VRAM Guidelines
  • Issue comment#7Vibecoder90002025-05-10 19:14
    How hard it is for Tencent to publish GGUF weights and demo Gradio app with GGUF for consumer GPUs? Sadly we are limited to NVIDIA until China brings Consumer GPUs with high vram
  • Issue comment#4frosty39072025-05-10 03:54
    only support one image as reference?

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 465 stars here means stars gained during the window, not the repo's star count.