Skip to content

FreeVideo v0.4.12

Choose a tag to compare

View all tags
@github-actions github-actions released this 17 Oct 15:59
· 209 commits to main since this release
v0.4.12
62f8900
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
FreeVideo

Download / 下载
Windows 10/11 · NVIDIA GPU FreeVideo.exe
macOS 14+ · Apple silicon FreeVideo-Mac-arm64.dmg
Linux x86_64 · NVIDIA GPU FreeVideo-Linux-x86_64.AppImage

Windows: run FreeVideo.exe, choose an existing ComfyUI folder or install a new one, then click Install & launch. Offline packages are on Quark and ModelScope.

macOS: open the DMG and drag FreeVideo.app into Applications. If macOS asks you to confirm the first launch, see the first-open steps.

Linux: make the AppImage executable (chmod +x FreeVideo-Linux-x86_64.AppImage) and open it, as on Windows. On a server or from a terminal, see Linux.

Windows： 运行 FreeVideo.exe，选择已有的 ComfyUI 或安装新的 ComfyUI，点击 安装并启动。离线包见夸克网盘或魔搭。

macOS： 打开 DMG，将 FreeVideo.app 拖入「应用程序」。首次打开如需确认，请参考首次打开步骤。

Linux： 给 AppImage 加上可执行权限（chmod +x FreeVideo-Linux-x86_64.AppImage）后打开，用法与 Windows 相同。在服务器或终端中使用请参考 Linux 说明。

What's new

Preview white-model scenes for Autopilot, LoRA compatibility checks and drag-to-reorder, clearer failure messages, deleting edits, a faster start for a busy machine, and a Medium level that samples two passes.

  • Autopilot: an optional white-model preview, with ready-made scenes, shows the scene before the video is made; the entry showcase has an official race-car example.
  • LoRA: a compatibility check before generating, and LoRAs can be reordered by dragging.
  • Failure messages: submission, queue and browser-extension failures share one layout and say what to do next; failed requests can be packaged with Export report or Submit to GitHub.
  • Editor: an edit can be deleted, an empty untouched edit is discarded, and reopening an edit restores its cover and shows its finishing stage; continuations are queued on the server.
  • Characters: each character carries a sheet of all views plus a face close-up; a picture that cannot be read is refused with a message on its own tile.
  • Medium: samples two passes from a larger first pass; Max runs all 20 steps without a cache.
  • Install: the GPU acceleration build has a visible share of the progress, the model line stays steady while groups download, and torch shows as Runtime components.
  • Downloads: before a download starts, the launcher, the page and the editor say what it is and how long it takes.
  • Guide: the first-run guide shows where to add reference images and first or last frames, the length field shows the part limit, and the Chinese interface links a user guide beside the guide in Settings.
  • Reusing reference pictures and clips when only the prompt changes skips encoding them again; on a busy machine the small CPU table work of loading runs on one thread.
  • Mac: a device that cannot continue a video refuses lengths over one block up front.
  • The feedback button opens the issue form.

更新内容

AI 全自动新增白模预览场景，LoRA 兼容性检查和拖动排序，失败提示更清楚，可以删除剪辑，繁忙的机器上启动更快，标准档改为两遍采样。

  • AI 全自动：可选的白模预览和现成场景，在生成视频前先看到场景；入口展示里有官方赛车示例。
  • LoRA：生成前做兼容性检查，LoRA 可以拖动排序。
  • 失败提示：提交、排队和浏览器扩展的失败用同一种版式，并说明下一步该做什么；失败的请求可以打包为导出报告或提交到 GitHub。
  • 编辑器：可以删除剪辑，没改动的空剪辑自动丢弃；重新打开剪辑会恢复封面并显示收尾阶段；续接在服务端排队。
  • 角色：每个角色带有全部视角的角色图和脸部特写；读不出来的图片会被拒绝，并在对应的格子上给出提示。
  • 标准档：从更大的第一遍起采样两遍；极致档跑满 20 步且不使用缓存。
  • 安装：GPU 加速组件的构建在进度中占有可见的份额，模型下载分组并行时进度行保持稳定，torch 显示为“运行组件”。
  • 下载：开始下载前，启动器、页面和编辑器会说明这是什么下载、需要多久。
  • 指引：首次指引新增一步，说明在哪里添加参考图和首帧、尾帧；时长栏直接显示分段上限；中文界面在设置里可以打开使用说明。
  • 只改提示词时，参考图和片段不再重复编码；机器繁忙时，加载过程中的小型 CPU 表运算改为单线程。
  • Mac：不能续接视频的设备会在一开始就拒绝超过一个分段的长度。
  • 反馈按钮直接打开问题表单。
sha256 
953f2c67cccd289251e01ef7d4610f10ffb22bd16dab329f99794eacedcb0f73  FreeVideo.exe
82f4388f98e4e7ccbc54040e0f75d435fd4f7a9f53394815412039cec42dfb86  FreeVideo-Mac-arm64.dmg
433a5b629924c88c6f0cfdd0838acd02d4d6246fde9d3f67aad6e24be3b2fb6f  FreeVideo-Linux-x86_64.AppImage

Discord · QQ Group / QQ 群 · WeChat Group / 微信群

Assets 9
Loading