Skip to content
OpenChatCut

OpenChatCut Pro · Desktop

A whole AI editing studio,installed on your machine.

Say it, the Agent cuts. Record and keep cutting. Works offline.

Download Pro v2.0.1See what it can do
  • 7-day trial, granted automatically on first launch
  • v2.0.1 · released Oct 27, 2026
  • macOS (Apple Silicon / Intel) · Windows
OpenChatCut Pro · Agent workspace
OpenChatCut Pro desktop app: Agent chat panel on the left, media and preview in the middle, multitrack timeline at the bottom

The OpenChatCut Pro desktop app.

71
built-in workflow skills
22
offline local models
4
transcription models to choose
≈20
LLM providers to plug in
7 days
free trial, no setup

Sections

One desktop app for the whole edit.

What the current version ships. Click through for details.

01Agent editing

Ask for an edit.The agent does the rest.

Chat mode to think out loud, Agent mode to work the timeline — every step runs the same commands as your hands, always undoable.

  • Captions, transitions, effects, music — say the word
  • Self-checks picture and loudness, then reports
  • Asks before guessing; every step undoable
  • Installs what it lacks: skills from repos, fonts online
  • Brand kit: captions, covers and MG align to your brand

Try asking

  • Remove the pauses in the second clip
  • Turn this long video into three highlight shorts
  • Line up the cuts with the beat of the music
  • Export this as a JianYing draft so I can finish it there
PRO / AGENT WORKSPACE
The OpenChatCut open-source editor, where an agent generates music and adjusts transitions on a multitrack timeline

02Transcript editing

Edit speechthe way you edit a document.

Footage becomes word-timestamped text: delete a sentence or mute a passage and the timeline follows.

PRO / TRANSCRIPT VIEW
OpenChatCut Pro transcript view: word-level timestamps — delete a sentence and the matching footage is cut
  • One click to drop pauses and filler words — preview first
  • Click a sentence to jump to its footage
  • Results saved with the project
  • Export SRT, or let AI rewrite

Local speech recognitionSettings → Local models

  • FunASR Nano

    Most accurate for Chinese, about 4× real time on a CPU. Chinese, English, Japanese, Korean and Cantonese.

    Chinese default

  • SenseVoice Small

    About 228 MB. Steadier with mixed Chinese and English and with English narration, and faster.

    Mixed languages

  • Whisper Small

    Covers 99 languages and can translate.

    Multilingual

  • Whisper Large v3 Turbo

    About 1 GB. The most accurate across languages; best for non-Chinese footage.

    Other languages

Cloud transcription (OpenAI, Deepgram, …) works too, with your own keys.

03Record & edit

Record and editwithout switching apps.

The record button lives in the editor toolbar. Takes land in the library, drop straight onto the timeline.

  • Microphone

    Voiceover and talking heads, check levels first

  • Screen

    Demos and tutorials, no extra tool

  • Camera

    On-camera, same project

A typical talking-head video, from recording to finished cut

  1. 01

    Record in place

    Recorded takes auto-import to the timeline

  2. 02

    Transcribe locally

    Local models turn audio into text

  3. 03

    Clean up the text

    Trim pauses and fillers, footage follows

  4. 04

    Finish the cut

    Add captions and music, export

PRO / RECORD IN PLACE
Recording inside the OpenChatCut Pro editor: mic, screen and camera captures land straight on the timeline

The Agent can also start and stop recording for you.

04Local models

Heavy lifting on your machine.Even offline.

Heavy work runs on your machine, offline included. Settings → Capabilities downloads everything this machine can run.

  • Speech recognitionFunASR · SenseVoice · Whisper ×2
  • Local voiceover9 offline voices
  • Matting & facesRVM ×2 · MediaPipe · YuNet · SFace
  • Denoise · upscale · separateDeepFilterNet3 · Real-ESRGAN · UVR
  • Vision understandingQwen3-VL 2B · Chinese CLIP
  • Text recognitionPP-OCR v4, straight to the timeline
  • Beats & alignmentbeat-this · wav2vec2 forced alignment ×2
  • Voice activity & musicSilero VAD · CLAP
  • Right-click a clip: matting, denoise, upscale, stabilize, separate
  • Resumable downloads, SHA-256 verified
  • Default model per capability; cloud optional
PRO / CAPABILITIES
OpenChatCut Pro local model manager: download by capability, resumable, SHA-256 verified

05Providers & MCP

Bring your own models.Connect the services you use.

Paste a key in Settings → Providers; logged-in Claude Code or Codex works as the model directly.

  • DeepSeek
  • OpenAI
  • Anthropic
  • Google
  • OpenRouter
  • Groq
  • Moonshot
  • xAI
  • and about a dozen more

Anything not listed plugs in via the OpenAI-compatible interface. Pair a vision model when the main one is text-only.

Image generation
OpenAI, Gemini, MiniMax, WaveSpeed, BytePlus, xAI
Voiceover TTS
OpenAI, MiniMax, ElevenLabs, Doubao, Fish Audio, Cartesia…
Video & music generation
Kling, MiniMax, Volcano Ark, Mureka, Sonilo…
Video download
yt-dlp: YouTube, Bilibili… managed copy, auto-updated
Stock & SFX
Pexels, Pixabay, Unsplash, Freesound, plus open museum & NASA archives

Each service bills on its own terms; not included in the Pro license.

Outside agent → your project

Open a project to Claude Code, Codex or Cursor

Settings → External Agents exposes an endpoint and token; configure your tool and it can edit the project.

  • Bearer token, regenerate anytime
  • Panel shows connected clients and call counts
  • Paste-ready setup instructions included

Add it in Claude Code (endpoint and token are in Settings → External agent)

claude mcp add --transport http openchatcut <endpoint> \
  --header "Authorization: Bearer <token>"
Your tools → the built-in agent

Connect local MCP servers to the built-in agent

Attach memory stores, databases or file systems as MCP servers the built-in Agent can call.

  • stdio MCP servers on this machine
  • Add and toggle in .agent/mcp-servers.json
  • External tools join the built-in editing tools

06The full editor

Beyond the AI,a complete editor.

Anything the Agent does, you can do by hand — and keep adjusting after it.

Open-source editor · LUT library
The OpenChatCut open-source editor with its LUT library, a vertical neon street preview and a multitrack timeline marked with effects and transitions
Illustration: the LUT library and timeline in the open-source editor. Pro's interface and features are those of the installed release.

Editing

  • Multitrack timeline, several sequences per project
  • Keyframes with custom Bézier easing
  • 9 mask types, 17 blend modes
  • Speed changes and speed curves
  • Native preview and proxy files

Picture

  • 87 effects
  • 42 transitions
  • Film looks, LUT import and AI color
  • Animated stickers and Lottie
  • Brand kit: colors, fonts and caption presets

Text and sound

  • 49 text animations
  • 57 caption presets, plus JianYing preset import
  • Karaoke caption templates
  • 38 bundled font families, plus your own
  • Music, sound effects and audio fades

Delivery

  • MP4, WebM, GIF, MP3 and WAV export
  • Hardware-accelerated encoding
  • 16:9, 9:16, 1:1 and 4:5 in one export
  • SRT and TXT captions
  • Pre-export check and export history

VSTwo editions, two ways to work

Build it yourself, or install and cut.

Open-source communityAGPL · $0

Read the source, make it yours.

Browser-first, AGPL. For developers who read source and build their own.

  • Open source, community-driven
  • Runs in the browser, or self-build
  • Bring your own keys
  • Local Whisper ASR, mic voiceover
Download open source
Pro desktop7-day free trial

Install and start cutting.

Tauri shell + Rust rendering core. Skills, local models, two-way MCP and signed installers, out of the box.

  • 71 finishing skills (27 in OSS), five-dimension self-review
  • 22 local models, offline
  • Screen and camera recording built in
  • Brand kit · yt-dlp downloads · self-installing agent
  • Two-way MCP to your AI toolchain
Download Pro

7-day trial starts automatically; after it ends editing, export and local models keep working.

?FAQ

A few things to know

Start

Your next cut starts with one sentence.

Download Pro, get 7 days free, finish your first cut today.

Download OpenChatCut ProEmail us for a license
  • Projects, media and keys stay local
  • Cloud powers are opt-in
  • Core features never lock after trial

Get started in three steps

  1. 01

    Download

    Pick the macOS or Windows installer — both signed.

  2. 02

    Get the trial

    7-day trial starts on first launch.

  3. 03

    Activate

    Editing stays free; email us for a license to unlock Agent and MCP.