ITG NowHome › AI glossary › Semantic search
What is Semantic search?
Semantic search is search that finds results by meaning, not only by the exact words you typed.
- Category Agents & tools
- Updated 20 Oct 2026
Semantic search, explained simply
Semantic search tries to understand what you mean. If you type "how to fix a leaky tap", it can also find pages about a dripping faucet.
It matters because normal keyword search misses good answers that use different words. Semantic search catches them.
For example, you can search your notes for "ideas for dinner" and find a page titled "easy recipes".
It usually works by turning text into numbers called embeddings, then comparing them.
On ITG Now, our free tool guides explain when a search tool uses meaning-based search, so you can find notes and files more easily.
Related terms
EmbeddingsVector databaseRAGLLM
Where you'll see it on ITG Now
Questions people ask about Semantic search
Semantic search vs keyword search: what is the difference?
Keyword search matches the exact words. Semantic search matches the meaning.
Is semantic search better than keyword search?
Not always. Exact names and codes are often easier to find with keyword search, so many tools use both.
How does semantic search work?
Text is turned into numbers that capture meaning. The system then finds the closest matches to your question.