Get This Tool
SynthCut
Pricing
- Model
- Free
Summary
Every video editor on the market is designed for a human hand on a mouse — which means when you want an AI model to execute a creative brief start-to-finish, you end up with a bot that can describe edits but cannot make them. SynthCut is built the other way around: the editor is the MCP server, and the AI client is the operator.
SynthCut exposes a full multi-track, frame-based video editor as an MCP server, so any MCP-compatible AI client — Claude Desktop being the documented example — drives real FFmpeg operations locally, fully offline. The architecture sidesteps the cloud-dependency and privacy concerns that come with hosted video AI tools. Where it breaks: the AI client is doing the driving, which means your workflow ceiling is whatever your MCP client can reason about and whatever tools SynthCut exposes. The project has 21 commits and 3 stars at time of scrape — early-stage by any measure. Teams that need a mature plugin ecosystem or a GUI-first fallback will hit that wall fast.
Bottom line: Pick SynthCut if you are building an AI-driven video pipeline that needs local FFmpeg execution and privacy guarantees — walk away if your team needs a stable, production-hardened editor with an established community behind it.
Hosted & API Pricing
The model is free to self-host. These are the creator's hosted/API options.Free
100 credits, no card required, sufficient for about two full videos
- Hosted video generation
- Captions, color grading, music sync
Pricing may have changed since last verified. Check the official site for current plans.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Full offline, local FFmpeg execution under GPL-3.0 license, so no frames leave your machine and no API billing accumulates — critical for privacy-sensitive productions or air-gapped environments.
- MCP server architecture lets an AI client iterate on edits programmatically — checking output, adjusting, re-running — rather than making a single opaque API call and hoping the result is right.
- Multi-format social media export from a single project is exposed as an AI-drivable operation, so a single agent instruction can produce platform-specific cuts without manual re-export cycles.
- Self-hosted with Windows installer and zip releases documented in the repository, which means deployment does not depend on a vendor staying solvent or keeping a SaaS instance running.
- Provider-agnostic at the MCP client level, so swapping the AI client driving the editor — or pointing it at a different underlying model — does not require changing the editor itself.
Cons
Sign in to edit- The tool only works if your AI client supports MCP — teams using OpenAI's API directly, LangChain agents without MCP integration, or non-MCP orchestration setups cannot drive SynthCut at all without building an adapter layer first.
- The repository has 21 commits and 3 stars at time of scrape, which means sparse documentation, no established community for debugging production issues, and real risk that an edge-case FFmpeg operation has no prior issue thread to reference — teams that hit an undocumented wall are on their own.
- There is no GUI-first fallback — if the AI client misinterprets a brief or produces a broken edit, a human cannot step in and correct it through the editor's own interface the way they could in Premiere or DaVinci Resolve; the correction loop runs back through the AI client, which adds latency and opacity to every fix.
- Teams that need feature depth — advanced color science, plugin ecosystems, collaborative review workflows, or broadcast-spec export profiles — will exhaust what SynthCut exposes as MCP tools quickly and end up building the gap themselves or switching to a mature editor with an AI integration layer bolted on.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- Windows
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- 2026-07-23T06:18:17.619Z
Best For
Who it's for
- Users with MCP-compatible AI clients
- Offline, privacy-focused editing workflows
- Developers integrating AI into video pipelines
What it does well
- AI-driven timeline editing via chat
- Local offline video production with FFmpeg
- Multi-format social media export from single project
- Captioning and color grading automation
- Semantic search and subject-aware reframing
Integrations
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare SynthCut
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Frequently Asked Questions
- Is SynthCut free?
- Yes — SynthCut is fully free to use. There is no paid tier.
- Is SynthCut open source?
- Yes. SynthCut is open source.
- Does SynthCut have an API?
- Yes. SynthCut exposes a developer API. See the official documentation at https://github.com/relo-video/synthcut for details.
- Can I self-host SynthCut?
- Yes. SynthCut supports self-hosting on your own infrastructure.
- What platforms does SynthCut support?
- SynthCut is available on: Windows.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
SynthCut reframes what a video editor is: instead of a canvas a human clicks through, it is an MCP server that an AI client queries and commands. The core workflow has an MCP-compatible AI client — the docs describe Claude Desktop as the reference setup — issuing editing instructions that SynthCut translates into real, local FFmpeg operations. Multi-track, frame-based edits happen on your machine, with no frames leaving to a third-party API.
The differentiating architectural choice is the MCP server exposure. Most AI video tools wrap a cloud pipeline and hand you back a rendered file. SynthCut gives the AI client direct, programmatic access to the editor state, which means the agent can iterate — checking what it produced, adjusting parameters, branching based on what the last step returned — rather than making a single blind call and waiting. The vendor describes use cases including captioning, color grading automation, semantic search, subject-aware reframing, and multi-format social media export from a single project.
This fits developer-led workflows where the goal is integrating AI into a video pipeline, not replacing a human editor with a shinier GUI. It also fits privacy-sensitive production: GPL-3.0 licensed, self-hosted, offline by design, with Windows installer and zip releases documented on the repository. Where it breaks is anywhere you need a production-stable tool with a track record — the repository is early-stage, community is minimal, and the issue tracker and contributor base reflect that. Teams accustomed to Premiere or DaVinci Resolve’s depth of feature coverage will find the gap significant.
Setup requires an MCP-compatible AI client configured to point at SynthCut’s MCP server; the repository includes a `claude_desktop_config.example.json` and a dedicated `MCP_CLIENT_SETUP.md` to reduce that friction. A separate hosted sibling product, relo.video, exists for teams that prefer a managed path — but SynthCut itself is the open-source, self-hosted, free variant.