Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.

A browser extension for Penetration Testing

Letta (formerly MemGPT) is the stateful agents framework with memory, reasoning, and context management.

Python SDK, Proxy Server (LLM Gateway) to call 100+ LLM APIs in OpenAI format - [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, Replicate, Groq]

Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

A Python package for fast and robust Image Stitching

OCR Benchmark

AirLLM 70B inference with single 4GB GPU

Standardized Serverless ML Inference Platform on Kubernetes

🔥 Sa2VA: Marrying SAM2 with LLaVA for Dense Grounded Understanding of Images and Videos

A minimal Python package for storing and retrieving text using chunking, embeddings, and vector search.

Generate server and client code from Swagger (OpenAPI 2.0) specification.

Generate modern Python clients from OpenAPI

Official inference library for Mistral models

PyTorch Implementation of "Distilling a Neural Network Into a Soft Decision Tree." Nicholas Frosst, Geoffrey Hinton., 2017.

MinIO is a high-performance, S3 compatible object store, open sourced under GNU AGPLv3 license.

The easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more!

Fully open reproduction of DeepSeek-R1

⚡ Create handwritten documents from text with a Neural Network!

A Python Package For System Identification Using NARMAX Models

A python library to build Model Trees with Linear Models at the leaves.

Python tool for converting files and office documents to Markdown.

Sonatype Nexus Repository Open-source codebase mirror

The Cloud Native Control Plane

Build Conversational AI in minutes ⚡️

🥤 RAGLite is a Python toolkit for Retrieval-Augmented Generation (RAG) with PostgreSQL or SQLite

A Faster LayoutReader Model based on LayoutLMv3, Sort OCR bboxes to reading order.

Official Implementation of CVPR24 highligt paper: Matching Anything by Segmenting Anything

Local models support for Microsoft's graphrag using ollama (llama3, mistral, gemma2 phi3)- LLM & Embedding extraction

Launch Dolphin from RetroArch/libretro