# llm-app | Pathway | Knowledge Base | Instagit

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, and more.

GitHub Stars: 56.2k

Repository: https://github.com/pathwaycom/llm-app

---

## Articles

### [How Pathway's Multimodal RAG Pipeline Extracts and Indexes Tables and Charts from PDFs Using GPT-4o](/pathwaycom/llm-app/how-does-pathways-multimodal-rag-pipeline-extract-and-index-tables-and-charts-from-pdfs-using-gpt-4o)

Discover how Pathway's multimodal RAG pipeline leverages GPT-4o to extract and index tables and charts from PDFs. Learn about its advanced data extraction and retrieval capabilities.

- Tags: how-to-guide
- Published: 2026-03-07

### [How to Migrate Existing Pinecone, Weaviate, or Qdrant Vector Database Implementations to Pathway](/pathwaycom/llm-app/how-do-i-migrate-existing-pinecone-weaviate-qdrant-vector-database-implementations-to-pathway)

Easily migrate Pinecone, Weaviate, or Qdrant vector databases to Pathway. Replace SDK calls with Pathway's VectorStoreServer and VectorStoreClient for faster ANN and hybrid search.

- Tags: migration-guide
- Published: 2026-03-07

### [Key Differences Between question_answering_rag, document_indexing, and adaptive_rag Templates in Pathway LLM-App](/pathwaycom/llm-app/what-are-the-key-differences-between-question-answering-rag-document-indexing-and-adaptive-rag-templates)

Understand the key differences between Pathway LLM-App's question_answering_rag, document_indexing, and adaptive_rag templates. Choose the best RAG solution for your needs.

- Tags: deep-dive
- Published: 2026-03-07

### [How to Implement Multi-Tenant Isolation in Pathway RAG Applications](/pathwaycom/llm-app/how-do-i-implement-multi-tenant-isolation-in-pathway-rag-applications)

Learn to implement multi-tenant isolation in Pathway RAG applications by deploying separate pipeline instances per tenant. Secure credentials persistence and network ports for robust isolation.

- Tags: how-to-guide
- Published: 2026-03-07

### [usearch Vector Indexing Performance at Scale: Handling Millions of Documents in Pathway LLM-App](/pathwaycom/llm-app/what-are-the-performance-characteristics-of-usearch-vector-indexing-at-scale-millions-of-documents)

Discover usearch vector indexing performance at scale. Pathway LLM-App handles millions of documents with sub-millisecond query latency using HNSW and SIMD acceleration.

- Tags: performance
- Published: 2026-03-07

### [How to Customize the REST API Response Schema in QASummaryRestServer: 3 Methods Explained](/pathwaycom/llm-app/how-do-i-customize-the-rest-api-response-schema-in-qasummaryrestserver)

Learn to customize the REST API response schema in QASummaryRestServer with 3 methods. Utilize request flags, the answer schema parameter, or subclassing for full control.

- Tags: how-to-guide
- Published: 2026-03-07

### [Best Practices for Deploying Pathway Apps on Kubernetes and Cloud Containers](/pathwaycom/llm-app/what-are-the-best-practices-for-deploying-pathway-apps-on-kubernetes-or-cloud-containers)

Deploy Pathway apps effortlessly on Kubernetes and cloud containers. Discover best practices for seamless application deployment and management. Learn how Pathway containers integrate smoothly.

- Tags: best-practices
- Published: 2026-03-07

### [How Pathway Handles Rate Limiting and Retry Logic for LLM API Calls](/pathwaycom/llm-app/how-does-pathway-handle-rate-limiting-and-retry-logic-for-llm-api-calls)

Learn how Pathway's LLM wrappers manage rate limiting and retry logic with configurable exponential backoff and fixed delay policies. Handle API errors and network failures seamlessly.

- Tags: how-to-guide
- Published: 2026-03-07

### [How to Implement Real-Time Alerting When Document Content Changes in Pathway](/pathwaycom/llm-app/how-do-i-implement-real-time-alerting-when-document-content-changes-on-connected-sources)

Implement real-time alerting for document changes with Pathway. Get instant Slack notifications when query answers change in connected sources like Google Drive using the drive_alert template.

- Tags: how-to-guide
- Published: 2026-03-07

### [Monitoring and Logging Options in LLM-App: Beyond MonitoringLevel.NONE](/pathwaycom/llm-app/what-monitoring-and-logging-options-are-available-beyond-monitoringlevel-none)

Explore LLM-App monitoring and logging beyond NONE. Discover BASIC and DETAILED levels for runtime metrics, Prometheus, and OpenTelemetry.

- Tags: deep-dive
- Published: 2026-03-07

### [How Pathway Handles Authentication with Google Drive, SharePoint, and S3 Data Sources](/pathwaycom/llm-app/how-does-pathway-handle-authentication-with-google-drive-sharepoint-and-s3-data-sources)

Learn how Pathway handles authentication with Google Drive, SharePoint, and S3 data sources using service accounts, client certificates, and AWS credentials for seamless data access.

- Tags: how-to-guide
- Published: 2026-03-07

### [How to Configure persistence_backend for Distributed Pathway Deployments: A Complete Guide](/pathwaycom/llm-app/how-do-i-configure-persistence-backend-for-distributed-pathway-deployments)

Configure persistence_backend for distributed Pathway deployments by setting it to shared storage like S3 GCS or PostgreSQL in app.yaml Ensure identical worker configurations for a coherent cluster cache

- Tags: how-to-guide
- Published: 2026-03-07

### [Pathway's Built-in usearch Vector Index Architecture vs. External Vector Databases](/pathwaycom/llm-app/what-is-the-architecture-of-pathways-built-in-usearch-vector-index-compared-to-external-vector-databases)

Explore Pathway's built-in usearch vector index architecture versus external databases. Discover how our in-process KNN engine eliminates network latency for faster LLM apps.

- Tags: architecture
- Published: 2026-03-07

### [How to Integrate Pathway with LangChain or LlamaIndex as a Retriever Backend](/pathwaycom/llm-app/how-do-i-integrate-pathway-with-langchain-or-llamaindex-as-the-retriever-backend)

Integrate Pathway with LangChain or LlamaIndex as a retriever backend using Pathway DocumentStore for real-time, continuously updated retrieval. Connect via PathwayVectorClient or PathwayVectorStore.

- Tags: how-to-guide
- Published: 2026-03-07

### [How Adaptive RAG Reduces Token Costs While Maintaining Accuracy in Pathway LLM Apps](/pathwaycom/llm-app/how-does-adaptive-rag-reduce-token-costs-while-maintaining-accuracy-compared-to-standard-rag)

Discover how Adaptive RAG slashes token costs for LLM apps. Get accurate results by retrieving only necessary documents, unlike standard RAG. Read more.

- Tags: deep-dive
- Published: 2026-03-07

### [How QASummaryRestServer Handles Concurrent Request Processing Under Load](/pathwaycom/llm-app/how-does-qasummaryrestserver-handle-concurrent-request-processing-under-load)

Discover how QASummaryRestServer efficiently handles concurrent requests under load using FastAPI and Pathway's parallel data-flow for seamless processing.

- Tags: performance
- Published: 2026-03-07

### [How to Implement Custom Data Source Connectors for Pathway Pipelines](/pathwaycom/llm-app/how-do-i-implement-custom-data-source-connectors-for-pathway-pipelines)

Learn to implement custom data source connectors for Pathway pipelines by subclassing BaseConnector or CustomConnector. Easily integrate your data into streaming applications with pw.io.custom.

- Tags: how-to-guide
- Published: 2026-03-07

### [Embedding Models Supported by pathway.xpacks.llm for Vector Indexing](/pathwaycom/llm-app/what-embedding-models-are-supported-by-pathway-xpacks-llm-for-vector-indexing)

Discover supported embedding models for vector indexing with pathway.xpacks.llm. Explore OpenAI and local SentenceTransformer options for your Pathway applications.

- Tags: api-reference
- Published: 2026-03-07

### [How DocumentStore Handles Real-Time Index Updates When Source Documents Change](/pathwaycom/llm-app/how-does-documentstore-handle-real-time-index-updates-when-source-documents-change)

Learn how DocumentStore streams changes from source documents, updating indexes in real-time without interrupting queries. Discover efficient text and vector index management.

- Tags: internals
- Published: 2026-03-07

### [How to Configure Hybrid Search Combining Vector Embeddings with Tantivy Full-Text Search](/pathwaycom/llm-app/how-do-i-configure-hybrid-search-combining-vector-embeddings-with-tantivy-full-text-search)

Learn to configure hybrid search combining vector embeddings and Tantivy full-text search. Combine USearch and BM25 for superior document retrieval in your Pathway LLM app. Get started today!

- Tags: how-to-guide
- Published: 2026-03-07

### [PersistenceMode.UDF_CACHING vs. Other Persistence Modes in Pathway](/pathwaycom/llm-app/what-is-the-difference-between-persistencemode-udf-caching-and-other-persistence-modes-in-pathway)

Understand PersistenceMode.UDF_CACHING vs other Pathway persistence modes. Learn how UDF_CACHING optimizes LLM calls for cost and speed, unlike full pipeline state persistence.

- Tags: deep-dive
- Published: 2026-03-07

### [How Pathway Live Data Synchronization Handles Incremental Updates from Connected Data Sources](/pathwaycom/llm-app/how-does-pathways-live-data-synchronization-handle-incremental-updates-from-connected-data-sources)

Pathway live data synchronization automatically propagates row level deltas from source connectors through downstream transformations without polling or full reloads.

- Tags: deep-dive
- Published: 2026-03-07