AI Agent Data & Runtime

High-performance cloud-native vector database for large-scale ANN and semantic retrieval

[EMNLP2025] "LightRAG: Simple and Fast Retrieval-Augmented Generation"

Free LLM API keys for GPT-5.4, Claude, DeepSeek, Gemini, Grok — copy, paste, use. Updated 3-5x daily. No credit card needed

TiDB is built for agentic workloads that grow unpredictably, with ACID guarantees and native support for transactions, analytics, and vector search. No data silos. No noisy neighbors. No infrastructure ceiling

AI gateway written in Go. Lightweight unified OpenAI-compatible API for OpenAI, Anthropic, Gemini, Groq, xAI & Ollama. LiteLLM alternative with observability, guardrails, streaming, costs and usage tracking

Database and retrieval layer for AI apps, supporting data use in semantic search and intelligent applications

ClickHouse® is a real-time analytics database management system

🍁 Sycamore is an LLM-powered search and analytics platform for unstructured data

Metering and Billing for AI, API and DevOps. Collect and aggregate millions of usage events in real-time and enable usage-based billing

📊 Cube Core is open-source semantic layer for AI, BI and embedded analytics

The best-benchmarked open-source AI memory system. And it's free

WFGY is heading toward WFGY 5.0 Polaris Protocol, a major open-source release for AI reasoning, RAG, agents, and real-world workflows. Includes Problem Map, Global Debug Card, WFGY 4.0, and the CFV Easter Egg

Lossless Context Management plugin for Hermes Agent — DAG-based context engine that never loses a message

The Postgres development platform. Supabase gives you a dedicated Postgres database to build your web, mobile, and AI applications

The live data layer for apps and AI agents. Create up-to-the-second views into your business, just using SQL

RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs

Event streaming platform for agentic AI. Continuously ingest, transform, and serve event streams in real time, at scale

The AI search platform

🚀2.3x faster than MinIO for 4KB object payloads. RustFS is an open-source, S3-compatible high-performance object storage system supporting migration and coexistence with other S3-compatible platforms such as MinIO and Ceph

A user-space file system for interacting with Google Cloud Storage

Nornicdb is a distributed low-latency, Graph+Vector, Temporal MVCC with all sub-ms HNSW search, graph traversal, and writes. Using Neo4j Bolt/Cypher and qdrant's gRPC means you can switch with no changes. Then, adding intelligent features like schemas, managed embeddings, LLM reranking+inferrence, GPU accel, Auto-TLP, Memory Decay, and MCP server

Open-source persistent memory for AI agent pipelines (LangGraph, CrewAI, AutoGen) and Claude. REST API + knowledge graph + autonomous consolidation

MCP Toolbox for Databases is an open source MCP server for databases

Simple unified access to frontier AI models

The Cloud-Native API Gateway and AI Gateway

Availability monitoring system for LLM relay services with real-time status dashboards

Real-time agent to agent communication

Fast hardware-isolated execution sandboxes for AI agents and untrusted code

A shared personal context file for every connected AI agent

Native Metal local LLM inference runtime optimized for Apple Silicon

Agent-ready web extraction and commerce intelligence delivered through structured APIs, CLI, and MCP

Persistent memory and knowledge layer for AI agents with MCP-native access to files, URLs, and context

MCP-native self-updating context layer shared across Claude, ChatGPT, Cursor, and everyday apps

Real-time external data marketplace for AI agents, connecting search, web, social, finance, and ecommerce data through one MCP plugin

Trading desk for LLM calls that optimizes token cost, latency, reliability, and quality through routing and arbitrage

Token compression gateway for coding agents that keeps context while reducing cost

DeepLake tool that gives Claude Code, Codex, Cursor, and other AI agents persistent shared team memory

Epsilla is an Agent-as-a-Service platform for creating and operating AI agents on enterprise knowledge

HydraDB is context infrastructure that makes AI applications intelligent, stateful, and memory-aware

Data infrastructure for AI applications with vector storage and retrieval workflows

Mem0 enables AI apps to continuously learn from past user interactions, enhancing their intelligence and personalization

Shared context and living knowledge graph for multiple AI agents

Usage runtime, metering, and entitlement governance platform for AI products

Shared, searchable external memory for every AI coding agent

Model-agnostic persistent memory infrastructure for production agents

ContextPool gives AI agents persistent memory between sessions by extracting useful context and feeding it back automatically

InstantDB is an open-source realtime backend that bundles database, auth, storage, and sync for AI-built and full-stack applications

Google's on-device ML/GenAI runtime framework and successor to TensorFlow Lite

Wafer uses autonomous AI agents to profile and optimize GPU inference performance across production stacks

Open workflow automation project that runs natural-language-described tasks automatically

Geekflare API provides scraping, screenshot, and RAG-ready extraction APIs to turn raw web pages into clean, structured data

Vector database capability for AI applications, semantic search, and intelligent data infrastructure

Local-first plug-in runtime and database for building products with extensible runtime slots

Backend, database, and user system platform for AI applications

On-premise-first AI data platform for computer vision datasets, annotation workflows, quality control, and AI operations

Claude Code context hygiene tool that caps verbose tool output and deduplicates same-session re-reads to keep context cleaner

OpenAI-compatible inference API for accessing 45+ frontier AI models with request-based pricing and privacy-focused handling

Distributed shared filesystem for AI agents and sandboxes

Native macOS AI database client with approval-first SQL generation

Trusted, permissioned, and verifiable data context for apps and AI agents

Search API returning source-linked web excerpts selected for agent queries

Open-source, local-first macOS workflow capture that turns reviewed screen interactions into structured AI training data

Local-first personal memory hub that shares user-controlled long-term context across Claude Code, Codex, OpenClaw, and other AI tools

Locally compress tool results, files, and session history for coding agents

Open-source desktop app for locally running and fine-tuning LLM, diffusion, and audio models

Open-source, AI-ready documentation framework built on Astro and Vite with Markdown as the source

Index Claude Code sessions into a per-project local SQLite store and automatically recall relevant context

Model-independent runtime for persistent memory, skill learning, self-verification, and cross-environment agent execution

Open-source multimodal data-pipeline SDK for robotics and physical AI with orchestration, versioning, provenance, and curation

One private gateway for cloud and self-hosted AI models

Give AI assistants secure access to internal company knowledge for real business work

Shared persistent memory for Claude Code, Cursor, Codex, and other AI tools

Search personal highlights, notes, and reading memories from Claude and ChatGPT

Control every AI request

Write and publish release notes from an AI assistant through MCP

On-device specialist models for speech, text and vision

Run persistent AI agents in dedicated Firecracker microVMs

Route coding requests by task complexity, cache state and subscription quota

Keep local work results and evidence for coding agents

Supply agents with web data through APIs, MCP and SDKs

Multimodal data platform for finding failures in physical AI models

Autonomous backend platform built for coding agents over MCP

Clean documents for AI locally while showing each removal

Shared memory for AI agents and human teammates

Rent a remote Mac mini configured for Claude Code

Data-center GPUs for agentic AI and long-context inference

Access public X data through a REST API and hosted MCP server

Transcribe local media and search Yap dictations through an on-device MCP server

Give AI agents access to live internet data

Route requests to the right LLM for cost, latency, and quality

Always-listening searchable memory for AI agents

Access more than 500 AI models through one API

An open-source system of record shared by AI agents and teams

Switch among 200+ models inside Claude Code and see compression and routing savings

Create and manage macOS, Linux, and Windows virtual machines on Apple Silicon Macs

Access OpenAI-, Anthropic-, and Gemini-compatible APIs through one key

Use 200+ video, image, audio, and chat models through a pay-per-request API and no-code Studio

Give coding agents persistent isolated Linux machines in the cloud

Optimize agent model usage and set spending controls while tasks run

Give AI agents persistent self-hosted browser sessions

Let AI agents access your own devices with approval