Clipto - Fully local, natural language search over terabytes of media

@zrimko  I appreciate you digging into the technical side of things.

Clipto re-indexes incrementally. When you add new footage, only that new file gets processed — not your entire library. And that's specifically designed to save you time. You don't have to sit through a full re-scan every time when drop a new clip into your folder. It just works quietly in the background, so you can keep editing or searching without interruption.😊

@trydoff Great question.

Yes, high-bitrate HEVC footage is something we encounter quite often, especially from GoPros, drones, and modern mirrorless cameras.

During indexing, Clipto normalizes media through our processing pipeline, so it can handle a wide range of formats and codecs, including H.264, H.265/HEVC, AV1, VP8/VP9, MPEG-2, ProRes, AAC, MP3, FLAC, WAV, and more.

In practice, we’ve successfully indexed and searched across libraries containing hundreds of gigabytes and even multiple terabytes of HEVC footage.

Processing speed will depend on your hardware, but HEVC itself is absolutely a first-class citizen in the workflow.

@mogabr Thank you for your kind words! We're very much looking forward to you experiencing Clipto.

@rumiza_shaikh Thanks! Performance has actually been one of the biggest engineering challenges for us.

We’ve spent a lot of time optimizing the entire stack, from model acceleration and inference efficiency to orchestration between different models and processing pipelines.

The goal is to make indexing large media libraries feel like a background task rather than something that takes over your machine.

That said, we’re definitely not done. There is still plenty of room for improvement, especially around memory footprint and resource utilization during large indexing jobs.

We’ve been shipping performance improvements continuously, and there are a few more significant optimizations currently in the pipeline.

Since you’re working with local-first systems yourself, I’d love to stay in touch and compare notes. This is one of those areas where there’s still a lot of unexplored territory.

@mooyan Yes. You can even use Clipto on the road, on a mountaintop set, in the desert, on a plane, or pretty much anywhere, without relying on the cloud. Give it a try, and we’d love to hear more of your feedback, thanks:)

@jianqiang_hao Thank you so much, really appreciate the support.

“Future of local file management” is exactly the direction we care about: making the files already sitting on your device easier to understand, search, and reuse without forcing everything into the cloud.

Would love to hear what you think after trying it with your own media.

@rydensun Really glad to hear that — and thanks for installing so quickly.This is exactly the kind of everyday friction we hoped to address. File hunting feels small in the moment, but when it happens day after day, it can quietly consume a surprising amount of time and attention.Would love to hear how Clipto works with your own media after you’ve had a chance to try it：）

@lisa_helicopter_l Thank you! That pain is exactly what pushed us to build Clipto.

Asset management is one of those invisible parts of creative work that takes a huge amount of time, but rarely gets talked about. If Clipto can help creators spend less time digging through folders and more time actually making things, that’s a big win for us.

@sam_alghaithi Great question! Supporting a wide range of hardware has actually been one of our biggest engineering challenges.

We approach it from two directions.

First, at the model layer, we use different model tiers optimized for different classes of machines. Depending on the available hardware, Clipto can choose between smaller and larger models to balance quality, speed, and resource usage.

Second, we’ve spent a lot of time optimizing the orchestration layer. Different workloads are scheduled differently depending on the machine’s capabilities.

On high-end systems, we can take advantage of more parallelism and process media much faster. On lower-powered machines, the priority shifts toward stability and responsiveness, making sure indexing doesn’t overwhelm the computer or interfere with normal work.

There’s still room for improvement, but a lot of the engineering effort has gone into making local AI practical on real-world hardware rather than assuming everyone has a top-spec machine.

@vikranth_reddy_bollam Thanks! We’d love to have you try it as well.

Pixel support is definitely something we’re interested in, but today our focus is on Mac. The product is fairly compute-intensive, and we wanted to start on a platform where we could deliver the best possible experience before expanding further.

We’ve spent a lot of effort optimizing for Apple Silicon and making large-scale media indexing practical on a local machine.

Mobile devices are absolutely on our radar, but for now we’re focused on continuing to improve the Mac experience first and then evaluating the best path to bring Clipto to other platforms.

Out of curiosity, what’s your primary use case on a Pixel? Notes, recordings, podcasts, or something else?