The Challenge
Traditional AI pipelines reprocess entire datasets when anything changes. Adding one document to 100,000 existing ones triggers a full re-embedding, re-indexing, and re-inference run — wasting hours of compute time and significant API costs.
The Solution
Pixeltable provides intelligent incremental updates. The system tracks which rows changed, identifies affected computed columns, and recomputes only the minimum necessary. Embedding indexes are updated incrementally too.
Implementation Guide
Step-by-step walkthrough with code examples
Automatic Incrementality
See how Pixeltable processes only what's changed.
1import pixeltable as pxt2from pixeltable.functions import openai34# Create table with expensive AI processing5docs = pxt.create_table('app.docs', {6 'document': pxt.Document,7 'title': pxt.String,8})910# Expensive embedding generation11docs.add_computed_column(12 embedding=openai.embeddings(13 docs.title, # Simplified for demo14 model='text-embedding-3-large'15 )16)1718docs.add_embedding_index('title', embedding=docs.embedding)1920# Initial load: processes all 100K documents21docs.insert([...hundred_k_documents])2223# Add 1 new document: processes ONLY the new one24docs.insert([{'document': 'new.pdf', 'title': 'Latest Report'}])25# ✅ 1 embedding generated, not 100,0012627# Update a title: recomputes ONLY that row's embedding28docs.update(29 {'title': 'Updated Report'},30 where=docs.title == 'Latest Report'31)32# ✅ 1 row recomputed, index updated incrementally
Key Benefits
Real Applications
Prerequisites
Performance
Learn More
Related Guides
Build a complete Retrieval-Augmented Generation pipeline with Pixeltable. Ingest documents, chunk text, generate embeddings, index for retrieval, and generate LLM answers — no vector database or orchestrator required.
Build AI agents with durable memory and tool-calling capabilities using Pixeltable and Model Context Protocol (MCP). Store conversations, images, and documents as queryable tables that agents can read from and write to.
Transcribe audio files at scale with Pixeltable and OpenAI Whisper. Automatic batching, error handling, and incremental processing.
Ready to Get Started?
Install Pixeltable and start building in minutes. One pip install, no infrastructure to manage.