Kernel Memory

Kernel Memory (KM) is a multi-modal AI Service specialized in the efficient indexing of documents and information through custom continuous data pipelines, with support for Retrieval Augmented Generation (RAG), synthetic memory, prompt engineering, and custom semantic memory processing.

KM supports PDF and Word documents, PowerPoint presentations, Images, Spreadsheets and more, extracting information and generating memories by leveraging Large Language Models (LLMs), Embeddings and Vector storage.

Utilizing advanced embeddings, LLMs and prompt engineering, the system enables Natural Language querying for obtaining answers from the information stored, complete with citations and links to the original sources.

Kernel Memory is designed for seamless integration with any programming language, providing a web service that can also be consumed as an OpenAPI endpoint for ChatGPT, web clients ready to use, and a Plugin for Microsoft Copilot and Semantic Kernel.

Kernel Memory (KM) and Semantic Memory (SM)

Kernel Memory (KM) is a service built on the feedback received and lessons learned from developing Semantic Kernel (SK) and Semantic Memory (SM). It provides several features that would otherwise have to be developed manually, such as storing files, extracting text from files, providing a framework to secure users’ data, etc. The KM codebase is entirely in .NET, which eliminates the need to write and maintain features in multiple languages. As a service, KM can be used from any language, tool, or platform, e.g. browser extensions and ChatGPT assistants.

Semantic Memory (SM) is a library for C#, Python, and Java that wraps direct calls to databases and supports vector search. It was developed as part of the Semantic Kernel (SK) project and serves as the first public iteration of long-term memory. The core library is maintained in three languages, while the list of supported storage engines (known as “connectors”) varies across languages.

Here’s comparison table:

Feature	Kernel Memory	Semantic Memory
Data formats	Web pages, PDF, Images, Word, PowerPoint, Excel, Markdown, Text, JSON, HTML	Text only
Search	Cosine similarity, Hybrid search with filters (AND/OR conditions)	Cosine similarity
Language support	Any language, command line tools, browser extensions, low-code/no-code apps, chatbots, assistants, etc.	C#, Python, Java
Storage engines	Azure AI Search, Elasticsearch, MongoDB Atlas, Postgres+pgvector, Qdrant, Redis, MSSQL Server, In memory KNN, On disk KNN. In progress: Azure Cosmos DB for MongoDB vCore, Chroma	Azure AI Search, Chroma, DuckDB, Kusto, Milvus, MongoDB, Pinecone, Postgres, Qdrant, Redis, SQLite, Weaviate
File storage	Disk, Azure Blobs, MongoDB Atlas, In memory (volatile)	-
RAG	Yes, with sources lookup	-
Summarization	Yes	-
OCR	Yes via Azure Document Intelligence	-
Security Filters	Yes	-
Large document ingestion	Yes, including async processing using queues (Azure Queues, RabbitMQ, File based or In memory queues)	-
Document storage	Yes	-
Custom storage schema	some DBs	-
Vector DBs with internal embedding	Yes	-
Concurrent write to multiple vector DBs	Yes	-
LLMs	Azure OpenAI, OpenAI, LLamaSharp via llama.cpp, LM Studio, Anthropic, Semantic Kernel connectors	Azure OpenAI, OpenAI, Gemini, Hugging Face, ONNX, custom ones, etc.
LLMs with dedicated tokenization	Yes	No
Cloud deployment	Yes	-
Web service with OpenAPI	Yes	-

Kernel Memory

Kernel Memory (KM) and Semantic Memory (SM)

Topics