Local natural language search across all your files
You know the file exists. Spotlight can't find it. Stella can. Just describe it, "my cheat sheet for the final exam," "the product spec from last week's meeting" and Stella finds it. Then drag it straight into Gmail, open it, or launch any app from the same bar. One shortcut for your whole Mac. Free. Private. No account. Bad filenames welcome.
We've lost hours of our lives to Spotlight returning nothing when we know the file exists. Naming a file untitled (4).docx or assignement_1(4).pdf guarantees you'll never see it again. Naming a PDF after a client doesn't help when you've forgotten the client's name.
So we built Stella. You describe the file the way you'd describe it to a coworker. "The pitch deck with the blue charts," "my tax return from last year," "that contract Sarah sent in February," and Stella finds it. One search bar across all your local folders.
Oh, and it's free.
A few things we care about: - 100% local. Your files never leave your device. No cloud, no telemetry on file contents, no account required. The embedding model (291MB, quantized) runs on your machine. - Fast. Filename matches return in ~5ms; semantic results land under 200ms. - Honest about quality. We benchmarked against Spotlight on 433 real queries — Stella was ~5x more accurate (0.83 vs 0.16 MRR@10), and Spotlight didn't surface a single file Stella missed. We'd rather show numbers than adjectives.
What's next: more file formats, a Windows build, and cloud storage integration (yes that includes Outlook, because Outlook search is so unbelievably bad).
We'd love your feedback, especially the queries where it fails. That's where we learn the most.
– The Stella team
About Stella on Product Hunt
“Local natural language search across all your files”
Stella launched on Product Hunt on June 1st, 2026 and earned 92 upvotes and 9 comments, placing #14 on the daily leaderboard. You know the file exists. Spotlight can't find it. Stella can. Just describe it, "my cheat sheet for the final exam," "the product spec from last week's meeting" and Stella finds it. Then drag it straight into Gmail, open it, or launch any app from the same bar. One shortcut for your whole Mac. Free. Private. No account. Bad filenames welcome.
On the analytics side, Stella competes within Mac, Productivity and Search — topics that collectively have 774.3k followers on Product Hunt. The dashboard above tracks how Stella performed against the three products that launched closest to it on the same day.
Who hunted Stella?
Stella was hunted by Senan Gaffori. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
For a complete overview of Stella including community comment highlights and product details, visit the product overview.
Senan Gaffori
Hey Product Hunt 👋
We've lost hours of our lives to Spotlight returning nothing when we know the file exists. Naming a file untitled (4).docx or assignement_1(4).pdf guarantees you'll never see it again. Naming a PDF after a client doesn't help when you've forgotten the client's name.
So we built Stella. You describe the file the way you'd describe it to a coworker. "The pitch deck with the blue charts," "my tax return from last year," "that contract Sarah sent in February," and Stella finds it. One search bar across all your local folders.
Oh, and it's free.
A few things we care about:
- 100% local. Your files never leave your device. No cloud, no telemetry on file contents, no account required. The embedding model (291MB, quantized) runs on your machine.
- Fast. Filename matches return in ~5ms; semantic results land under 200ms.
- Honest about quality. We benchmarked against Spotlight on 433 real queries — Stella was ~5x more accurate (0.83 vs 0.16 MRR@10), and Spotlight didn't surface a single file Stella missed. We'd rather show numbers than adjectives.
What's next: more file formats, a Windows build, and cloud storage integration (yes that includes Outlook, because Outlook search is so unbelievably bad).
We'd love your feedback, especially the queries where it fails. That's where we learn the most.
– The Stella team