-
Notifications
You must be signed in to change notification settings - Fork 0
[SUBMISSION] Medha #20
Copy link
Copy link
Closed
Labels
approved-submissionLabel used to trigger workflow and create new pr: Issue to PRLabel used to trigger workflow and create new pr: Issue to PRcheck-submissionLabel used to trigger workflow in validation mode: Issue to PRLabel used to trigger workflow in validation mode: Issue to PR
Description
Metadata
Metadata
Assignees
Labels
approved-submissionLabel used to trigger workflow and create new pr: Issue to PRLabel used to trigger workflow and create new pr: Issue to PRcheck-submissionLabel used to trigger workflow in validation mode: Issue to PRLabel used to trigger workflow in validation mode: Issue to PR
Project Name
Medha
Description
Medha is an asynchronous, high-performance semantic cache library designed specifically for Text-to-Query systems.
Unlike traditional key-value caches that require exact string matches, Medha understands that "Show me the top 5 users" and "List the first five users" are the same question. It intercepts these queries and returns pre-calculated database queries (SQL, Cypher, etc.), bypassing the expensive and slow LLM generation step.
Category
library
Country
Italy
Platform
GitHub
Repository URL
https://github.com/ArchAI-Labs/medha
License
Apache-2.0
Language
Python
Owner Name
ArchAI Labs
Owner Type
community
Owner Description
ArchAI-Labs is an innovation hub dedicated to bridging the gap between complex codebases and human understanding. We build autonomous agents and intelligent tools designed to automate documentation and enhance software observability.
Owner Website URL
https://github.com/ArchAI-Labs
Is a Startup
No
Documentation URL
No response
Tags
agent memory, semantic cache, text2query
Terms & Conditions