Frequently Asked Questions
Clarifying confusion around RAG vs keyword search, and how Locaddit works.
What is RAG and how is it different from keyword search?
RAG (Retrieval-Augmented Generation) is a technique that combines semantic understanding with information retrieval. Unlike keyword search which only matches exact words, RAG understands the meaning behind your query. For example, if you search for "cheap headphones," RAG will also find discussions about "budget audio gear" or "affordable earbuds" because it understands these are semantically similar concepts.
Why can't I just use Reddit's built-in search?
Reddit's native search is keyword-based and has significant limitations. It often misses relevant content because it requires exact word matches, doesn't understand synonyms or context, and has limited indexing. Locaddit uses semantic search to find discussions even when they use different terminology than your query.
How does Locaddit avoid hallucinations?
Every answer Locaddit provides is grounded in actual Reddit content. We use RAG to retrieve relevant threads first, then generate responses based only on that retrieved context. All answers include citations linking back to the original Reddit posts, so you can verify the information yourself.
What subreddits are indexed?
Locaddit indexes content from across Reddit, including both popular and niche subreddits. We continuously crawl posts and comments to keep our index up-to-date. This includes everything from major communities to specialized hobbyist forums.
How recent is the data?
Our indexing pipeline runs continuously to capture new Reddit content. The exact freshness depends on our crawl frequency, but we aim to include recent discussions within hours of publication. Historical data goes back as far as our indexing has been running.
Can I search for specific subreddits?
Currently, Locaddit searches across all indexed subreddits to give you the most comprehensive results. However, results include the source subreddit, so you can filter or focus on specific communities if needed. Subreddit-specific filtering may be added in future updates.
Is Locaddit free to use?
Locaddit is currently in development. Pricing and availability will be announced closer to launch. We're committed to making Reddit's knowledge more accessible, and we'll share details about our access model as we approach release.
How accurate are the search results?
Locaddit uses state-of-the-art embedding models to ensure high-quality semantic matching. Results are ranked by relevance, and we show similarity scores to help you understand how closely each result matches your query. Since all results link back to original Reddit threads, you can always verify the information.
Can I use Locaddit for research or commercial purposes?
Locaddit provides access to publicly available Reddit content. However, you should respect Reddit's terms of service and content licensing. For commercial use, please review Reddit's API terms and content policies. We recommend citing original Reddit threads when using information from Locaddit.
How do I report incorrect or problematic results?
We're committed to improving Locaddit's accuracy and safety. If you encounter incorrect information, problematic content, or technical issues, please contact us through our support channels. Your feedback helps us refine our models and filtering systems.