Documentation
In-browser semantic search
Replace keyword search with semantic understanding. Moss finds answers based on meaning, not just matching words. Runs on your infrastructure, offline-capable, sub-10ms.
<10ms
query response time
100%
offline capable
0
external data transfers
The Bottom Line
Replace keyword search with semantic understanding. Moss finds answers based on meaning, not just matching words. Runs on your infrastructure, offline-capable, sub-10ms.
The Problem
Keyword search fails your users
Your users search for "how do I reset my password" and keyword search returns nothing because your docs say "credential recovery." Traditional search requires exact term matches. Semantic search understands meaning. But most semantic search solutions require sending your documentation and user queries to a third-party cloud service. For internal knowledge bases with proprietary information, that is a security risk you do not need to take.
Solution
How Moss Solves This
Search by meaning, not keywords
Moss uses vector embeddings to understand the semantic meaning of queries. Users find answers even when they use different words than your documentation.
Runs on your infrastructure
Deploy on your servers, in the browser, or on-device. Your documentation and user queries never leave your controlled environment.
Works offline
Once the index is loaded, all queries run in-memory. No network connection required. Ideal for internal tools, field service apps, and airgapped environments.
from moss import MossClient, QueryOptions
client = MossClient(PROJECT_ID, PROJECT_KEY)
# Create an index with your help articles
await client.create_index("help-center", help_articles)
await client.load_index("help-center")
# Users search by meaning, not keywords
results = await client.query(
"help-center", "how do I reset my password",
QueryOptions(top_k=5)
)
# Returns "credential recovery" docs — semantic match
for doc in results.docs:
print(f"[{doc.score:.3f}] {doc.text}")FAQ