Kikiola is a high-performance vector database written in Go. It efficiently stores, indexes, and searches for vectors, making it suitable for similarity search, recommendation systems, artificial intelligence, and machine learning applications.
- Tensor Compression
- Support multilingual embedding
- Support for high-dimensional vectors
- Handles concurrency and multiple writes
- Simple and intuitive API for easy integration
- Indexing techniques for fast similarity search
- Embedding Reranking with a relevance score
- Fast and efficient vector storage and retrieval
- Text embedding support for text-based queries
- Scalable architecture for handling large datasets
- Distributed Storage: multiple nodes or shards for scalability
- Objects (e.g., document, image, audio, video, or any other file type)
To run Kikiola, ensure that you have Go installed on your system. Then, follow these steps:
- Clone the Kikiola repository:
git clone https://github.com/0xnu/kikiola.git
- Navigate to the project directory:
cd kikiola
- Build the project:
go build ./...
- Run the Kikiola server:
go run cmd/main.go
The Kikiola server will start running on http://localhost:3400
.
To test Kikiola, ensure that you have Go installed on your system. Then, follow these steps:
go test ./...
- Usage
- Docker
- Benchmark
- Quay
- JFrog
- GitLab
- Microsoft Azure
- Amazon Web Services (AWS)
- Google Cloud Platform (GCP)
- Generate and Store Embeddings - Documents and Images
- Generate and Store Embeddings - Genome Sequence
- Generate and Store Embeddings - Multilingual
- Generate and Store Embeddings - Hugging Face 🤗
- Generate and Store Embeddings - SEC Form 10-K - Be a responsible Human and use the EDGAR API. 😎
- Generate and Store Embeddings - Videos
- Generate and Store Embeddings - Audios
This project is licensed under the MIT License.
(c) 2024 Finbarrs Oketunji.