Neura Market
llm-app logo

llm-app

Free

Ready-to-run cloud templates for RAG, AI pipelines, and enterprise search with live data. 🐳Docker-friendly.⚡Always in sync with Sharepoint, Google Drive, S3, Kafka, PostgreSQL, real-time data APIs, a

Search ToolsFreeFree tier
Type
Open Source
Visit Website GitHub (59.9k stars)

About llm-app

Pathway Live Data Framework Templates provide ready-to-deploy YAML and Python templates for building scalable, real-time RAG and ETL pipelines. The templates enable high-accuracy retrieval-augmented generation (RAG) using live data from sources such as Google Drive, SharePoint, S3, Kafka, PostgreSQL, and local file systems. Key offerings include basic question-answering RAG apps, Adaptive RAG (reducing token cost up to 4x), private RAG with local models (Mistral/Ollama), multimodal RAG with GPT-4o for PDFs and slide decks, live document indexing services compatible with LangChain/LlamaIndex, and structured data pipelines for financial reports. The framework features built-in connectors for 300+ data sources, incremental stream processing, sub-millisecond query latency, and supports Python/SQL APIs. Deployment is self-hosted via pip, poetry, or Docker, with community (free), scale (free up to 16GB RAM), and enterprise tiers available.

Key Features

Ready-to-deploy YAML and Python templates for RAG and ETL pipelines
Live data indexing from Google Drive, SharePoint, S3, Kafka, PostgreSQL, and 300+ data sources
Adaptive RAG technique reduces token cost up to 4x while maintaining accuracy
Multimodal RAG with GPT-4o for PDFs, slides, and unstructured documents
Private RAG support with local models (Mistral, Ollama) for fully offline operation
Real-time document vector store / retriever compatible with LangChain and LlamaIndex
REST API endpoints with sub-millisecond latency for querying and real-time features
Incremental stream processing (join, filter, group-by, temporal joins) in Rust engine
Built-in connectors for Kafka, PostgreSQL, HTTP, Redpanda, Slack, file systems, and more
Supports Python and SQL programming APIs with compile-time validation

Pros & Cons

Pros
  • Open-source (BSL 1.1) with free community tier supporting up to 16GB RAM
  • True streaming data processing engine in Rust for high performance and low latency
  • Live sync with data sources ensures answers are always up-to-date without manual reindexing
  • Docker-friendly deployment – ready-to-run containers for quick prototyping
  • Adaptive RAG reduces token costs up to 4x while maintaining accuracy
  • Supports a wide range of input connectors (files, cloud storage, databases, message queues)
  • Compatible with major LLM providers and local models for flexibility
Cons
  • Requires self-hosting; no fully managed cloud offering in the free tier
  • BSL 1.1 license may restrict commercial use without a paid license
  • Setup and configuration may require technical expertise (Docker, YAML, Python)
  • Community support only for free tier; business support requires Scale or Enterprise license
  • Limited to Pathway framework – not a general-purpose AI platform

Best For

Question-answering on live documents (PDF, DOCX, PPTX) from cloud or local storageFinancial report analysis: extract tables, charts, and answer natural language queriesSlide deck search: multi-modal indexing of PowerPoint and PDF presentationsPrivate RAG for sensitive data using local LLMs (Mistral, Ollama) without external API callsLive document indexing as a vector store service for frontend or LangChain/LlamaIndex backendsReal-time ETL pipelines for Kafka, Delta Lake, and database synchronizationSocial media sentiment analysis with streaming data

Alternatives to llm-app

cognee

The memory for your AI Agents in 6 lines of code

databend

Data Agent Ready Warehouse : One for Analytics, Search, AI, Python Sandbox. — rebuilt from scratch. Unified architecture on your S3.

awesome-llm-apps

100+ AI Agent & RAG apps you can actually run — clone, customize, ship.

claude-mem

A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future ses

deeplake

Deeplake is AI Data Runtime for Agents. It provides serverless postgres with a multimodal datalake, enabling scalable retrieval and training.

anything-llm

The all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.

FAQ

What is the Pathway Live Data Framework?
Pathway Live Data Framework is a scalable, real-time data processing framework for building AI/ML applications that continuously sync with live data sources. It provides a high-performance streaming engine written in Rust, with Python and SQL APIs, and a set of templates for RAG and ETL pipelines.
What templates are available?
Templates include basic question-answering RAG, Adaptive RAG (token-saving), private RAG with local models (Mistral/Ollama), multimodal RAG with GPT-4o, live document indexing as a vector store, slide AI search, and financial report structuring pipelines.
Is the framework free?
Yes, the Community edition is free and open under BSL 1.1, supporting up to 8GB RAM and 4 cores. Scale edition is free up to 16GB RAM and 4 cores. Enterprise edition offers unlimited resources and additional features with a paid license.
Can I run the templates locally?
Yes, templates are designed for self-hosted deployment using pip, poetry, or Docker. They can run on your own machine or cloud infrastructure.
What data sources are supported?
The framework provides connectors for Kafka, PostgreSQL, HTTP/JSON Lines, Redpanda, Logstash, Slack, file systems, Google PubSub, and more. Enterprise connectors include SharePoint, Delta Table, Iceberg, BigQuery, Elasticsearch, and QuestDB.