KEYWORDS ARE
A LIE.
Traditional search is a graveyard of metadata. It doesn't care about the soul of the story—it only cares if the title matches your typo.
The Popularity Trap
Old search engines only show you what's trending, burying the hidden gems you actually need.
The Keyword Curse
Tags are limited. Feelings aren't. Why can't you search for 'movies that feel like a cold morning'?
The Scroll Fatigue
Spent 45 minutes looking and 0 minutes watching? That's an algorithm failure, not a you problem.
Vibe-First Search
Stop typing titles. Start typing feelings. Search for "Lush atmospheric nostalgia," "Neon-soaked revenge," or "Heartfelt rural coming-of-age." Our semantic engine understands the DNA of story.
- Vector-based thematic retrieval
- Natural language understanding
- Mood-aware results
Your Cinematic
Signature
Connect your Letterboxd diary to unlock Neural Search. SBTXT analyzes your 5-star ratings to map your unique Taste DNA.
Taste DNA
Search using your personalized neural signature.
Watchlist Scan
Find exactly what to watch next in your own list.
Auto-Filter
Automatically hide every movie you've already seen.
Affinity Bonus
Your highest ratings carry the most weight.
Powered by the World's Movie Database
SBTXT is built upon the robust infrastructure of TMDB. We ingest millions of data points—from classic filmography to upcoming blockbusters—to ensure your discovery journey is always backed by the most comprehensive movie library on Earth.
Data Provided by TMDBGalaxy
Escape the grid. We've mapped over 100,000 films into a high-dimensional neural matrix. By measuring Raw Vector Similarity, we've created a universe where proximity is meaning.
Neural Embeddings
Every film distilled into a 768-dimensional mathematical signature representing its cinematic soul.
Spatial Projection
Advanced UMAP algorithms compress infinite variables into a navigable 3D star-field where math meets intuition.
Raw Similarity
Distance is data. If two stars are close, they share the same emotional, stylistic, and thematic DNA.