---
title: "Pixeltable notebooks and cookbooks"
description: "103 runnable Pixeltable notebooks and recipes: tutorials, cookbooks by topic, and provider guides."
url: "https://www.pixeltable.com/notebooks"
---

# Pixeltable notebooks and cookbooks

Most notebooks open in Google Colab; a recipe is a docs page with code and no notebook. They are the tutorials, cookbooks and provider guides on docs.pixeltable.com. A notebook that calls a model provider asks for its API key in a cell.

## Tutorials

- [10-Minute Tour](https://www.pixeltable.com/notebooks/ten-minute-tour.md): Welcome to Pixeltable! In this tutorial, we'll survey how to create tables, populate them with data, and enhance them with built-in and user-defined transformations and AI operations.
- [Tables and Data Operations](https://www.pixeltable.com/notebooks/tables-and-data-operations.md): Create and manage tables, insert, update and delete data, and filter and select rows with where(), select() and order_by().
- [Computed Columns](https://www.pixeltable.com/notebooks/computed-columns.md): This guide introduces one of Pixeltable's most essential and powerful concepts: computed columns.
- [Queries and Expressions](https://www.pixeltable.com/notebooks/queries-and-expressions.md): Expressions are the basic building blocks of Pixeltable.
- [UDFs in Pixeltable](https://www.pixeltable.com/notebooks/udfs-in-pixeltable.md): Pixeltable comes with a library of built-in functions and integrations, but sooner or later, you'll want to introduce some customized logic into your workflow.
- [Embedding Indices](https://www.pixeltable.com/notebooks/embedding-indexes.md): Create an embedding index, query it with similarity search and keep it current as data changes. Covers several indexes on one column, custom embedding UDFs and precomputed embeddings.
- [External Files](https://www.pixeltable.com/notebooks/external-files.md): In Pixeltable, all media data (videos, images, audio) resides in external files, and Pixeltable stores references to those.

## Core Operations

- [Your Backend for Multimodal AI Applications](https://www.pixeltable.com/notebooks/multimodal-backend.md): Every multimodal AI app needs the same five things: store media, run models, index embeddings, serve endpoints, version everything.
- [Configure API keys for AI services](https://www.pixeltable.com/notebooks/workflow-api-keys.md): Set up API credentials for OpenAI, Anthropic, and other AI providers so Pixeltable can access them.
- [Get fast feedback on transformations](https://www.pixeltable.com/notebooks/dev-iterative-workflow.md): You need to iterate on transformation logic before running it on your entire dataset—especially for expensive operations like API calls or model inference.
- [Working with the Pixeltable CLI](https://www.pixeltable.com/notebooks/working-with-cli.md): You defined a multimodal pipeline in Python and now need to inspect tables, debug computed columns, roll back changes, and expose HTTP endpoints without writing more application code.
- [Extract fields from LLM JSON responses](https://www.pixeltable.com/notebooks/workflow-json-extraction.md): Parse and access specific fields from structured JSON responses returned by language models.
- [Add unique identifiers to your tables](https://www.pixeltable.com/notebooks/workflow-uuid-identity.md): Generate UUIDs for automatic row identification.
- [Join tables to combine data](https://www.pixeltable.com/notebooks/query-join-tables.md): Combine data from multiple tables using inner, left, and cross joins.
- [Track changes and revert to previous versions](https://www.pixeltable.com/notebooks/version-control-history.md): Undo mistakes, audit changes, and create point-in-time snapshots of your data.
- [Create custom aggregate functions (UDAs)](https://www.pixeltable.com/notebooks/custom-aggregates-uda.md): Build reusable aggregation logic for group-by queries and analytics.
- [Custom Iterators](https://www.pixeltable.com/notebooks/custom-iterators.md): An iterator in Pixeltable is a function that expands a single input row into multiple output rows.
- [Split data into multiple rows with iterators](https://www.pixeltable.com/notebooks/data-split-rows.md): Transform a single document, video, image, or audio file into multiple rows for granular processing.
- [Time Zones](https://www.pixeltable.com/notebooks/time-zones.md): Because typical use cases involve datasets that span multiple time zones, Pixeltable strives to be precise in how it handles time zone arithmetic for datetimes.

## Agents

- [Agentic Patterns](https://www.pixeltable.com/notebooks/agentic-patterns.md): Build the six architectural patterns of agentic AI (prompt chaining, routing, parallelization, tool use, evaluator-optimizer, orchestrator-worker) and the ReAct and planning strategies as Pixeltable tables and computed columns.
- [Build a RAG pipeline](https://www.pixeltable.com/notebooks/pattern-rag-pipeline.md): Create a retrieval-augmented generation system that answers questions using your documents as context.
- [LLM as judge](https://www.pixeltable.com/notebooks/pattern-llm-judge.md): Score model output with a second model, as a chain of computed columns
- [Document Indexing and RAG](https://www.pixeltable.com/notebooks/rag-demo.md): In this tutorial, we'll demonstrate how RAG operations can be implemented in Pixeltable. In particular, we'll develop a RAG application that summarizes a collection of PDF documents and uses ChatGPT to answer questions about them.
- [RAG Operations in Pixeltable](https://www.pixeltable.com/notebooks/rag-operations.md): In this tutorial, we'll explore Pixeltable's flexible handling of RAG operations on unstructured text.
- [Build an agent with memory](https://www.pixeltable.com/notebooks/pattern-agent-memory.md): Create an AI agent that remembers important information across conversations.
- [Use a table pipeline as a reusable function](https://www.pixeltable.com/notebooks/pattern-table-as-udf.md): Convert a table with computed columns into a callable function for multi-agent workflows and pipeline composition.
- [Look up structured data with retrieval UDFs](https://www.pixeltable.com/notebooks/pattern-data-lookup.md): Create lookup functions that query tables by key—for customer records, product catalogs, or financial data.
- [Use tool calling and MCP servers with LLMs](https://www.pixeltable.com/notebooks/llm-tool-calling.md): Enable LLMs to call functions and tools, then execute the results automatically.

## Text

- [Summarize text with LLMs](https://www.pixeltable.com/notebooks/text-summarize.md): Generate concise summaries of long text, articles, or documents using large language models.
- [Translate text between languages](https://www.pixeltable.com/notebooks/text-translate.md): Automatically translate content into multiple languages using LLMs.
- [Extract named entities from text](https://www.pixeltable.com/notebooks/text-extract-entities.md): Identify and extract people, organizations, locations, dates, and other entities from text using LLMs.
- [Ingest a web page](https://www.pixeltable.com/notebooks/doc-ingest-website.md): Insert a URL as a document, then compare chunking strategies on it
- [Split documents into chunks for RAG](https://www.pixeltable.com/notebooks/doc-chunk-for-rag.md): Break PDFs and documents into searchable chunks for retrieval-augmented generation (RAG) pipelines.
- [Extract text from PowerPoint, Word, and Excel files](https://www.pixeltable.com/notebooks/doc-extract-text-from-office-files.md): Transform office documents into searchable, analyzable text data.

## Images

- [Detect objects in images and videos with YOLOX](https://www.pixeltable.com/notebooks/yolox-object-detection.md): Run YOLOX object detection in Python on images and video frames, then query and draw the detected objects.
- [Detect objects in images](https://www.pixeltable.com/notebooks/img-detect-objects.md): Automatically identify and locate objects in images using YOLOX object detection models.
- [Compare object detection and panoptic segmentation](https://www.pixeltable.com/notebooks/img-detection-vs-segmentation.md): Understand when to use bounding boxes versus pixel-level masks for image analysis.
- [Visualize object detections](https://www.pixeltable.com/notebooks/img-visualize-detections.md): Draw bounding boxes on images to visualize object detection results.
- [Promptable segmentation with SAM 3](https://www.pixeltable.com/notebooks/img-promptable-segmentation.md): Segment every object matching a concept (a short text phrase or a bounding box) using Meta's Segment Anything Model 3.
- [Generate captions for images](https://www.pixeltable.com/notebooks/img-generate-captions.md): Automatically create descriptive captions for images using AI vision models.
- [Transform images with AI-powered editing](https://www.pixeltable.com/notebooks/img-image-to-image.md): You have a batch of images that need AI-powered transformations—like turning photos into paintings, adding stylistic effects, or modifying content based on text prompts.
- [Analyze images in batch with AI vision](https://www.pixeltable.com/notebooks/vision-batch-analysis.md): Run the same AI prompt against multiple images automatically, without writing loops or managing API calls.
- [Extract structured data from images](https://www.pixeltable.com/notebooks/vision-structured-output.md): Use AI vision to extract JSON data from receipts, forms, documents, and other images.
- [Add watermarks to images](https://www.pixeltable.com/notebooks/img-add-watermarks.md): You need to add watermarks to hundreds of different images to protect copyright, add branding, or mark drafts.
- [Adjust image opacity](https://www.pixeltable.com/notebooks/img-adjust-opacity.md): You need to make hundreds of images semi-transparent for backgrounds, overlays, or watermarks.
- [Apply image filters](https://www.pixeltable.com/notebooks/img-apply-filters.md): You need to apply filters to hundreds of images—blur, sharpen, edge detection, and other enhancements.
- [Adjust image brightness and contrast](https://www.pixeltable.com/notebooks/img-brightness-contrast.md): You need to fix inconsistent lighting across hundreds of images—adjusting brightness, contrast, and color saturation.
- [Transform images with PIL operations](https://www.pixeltable.com/notebooks/img-pil-transforms.md): You need to resize, rotate, crop, or convert hundreds of images—and keep track of all the transformed versions.
- [Convert color images to grayscale](https://www.pixeltable.com/notebooks/img-rgb-to-grayscale.md): You need to convert color images to grayscale for analysis, preprocessing, or model inputs that require single-channel images.

## Audio

- [Transcribe speech and identify speakers with WhisperX](https://www.pixeltable.com/notebooks/whisperx-transcription.md): Transcribe speech with WhisperX in Python, identify speakers, and search the transcript with word-level timestamps.
- [Transcribe audio files with Whisper](https://www.pixeltable.com/notebooks/audio-transcribe.md): Convert speech to text locally using OpenAI's open-source Whisper model—no API key needed.
- [Transcribing and Indexing Audio and Video in Pixeltable](https://www.pixeltable.com/notebooks/audio-transcriptions.md): In this tutorial, we'll build an end-to-end workflow for creating and indexing audio transcriptions of video data.
- [Summarize podcasts and audio](https://www.pixeltable.com/notebooks/audio-summarize-podcast.md): Transcribe audio files and generate summaries automatically using Whisper and LLMs.
- [Generate Podcast Chapters](https://www.pixeltable.com/notebooks/audio-podcast-chapters.md): Generate YouTube-style chapter timestamps for podcast episodes. WhisperX detects where speech starts and stops, then an LLM summarizes each segment into a chapter title.
- [Convert text to speech](https://www.pixeltable.com/notebooks/audio-text-to-speech.md): Generate natural-sounding audio from text using OpenAI's text-to-speech models.
- [Extract audio from video](https://www.pixeltable.com/notebooks/audio-extract-from-video.md): Extract the audio track from video files for transcription, analysis, or processing.

## Video

- [Extract frames from videos](https://www.pixeltable.com/notebooks/video-extract-frames.md): Pull frames from video files at specified intervals for analysis, thumbnails, or training data.
- [Object Detection in Videos](https://www.pixeltable.com/notebooks/object-detection-in-videos.md): Detect objects in a video frame by frame: Pixeltable extracts the frames, runs YOLOX models on them and reassembles the frames into a video. Compare detection models and evaluate them against ground truth.
- [Detect scene changes in videos](https://www.pixeltable.com/notebooks/video-scene-detection.md): Automatically find scene cuts, transitions, and fades in video files.
- [Generate thumbnails from videos](https://www.pixeltable.com/notebooks/video-generate-thumbnails.md): Automatically create preview thumbnails from video files at specific timestamps or intervals.
- [Add text overlays to videos](https://www.pixeltable.com/notebooks/video-add-text-overlay.md): Burn text, captions, or watermarks directly into video files.
- [Generate videos with AI](https://www.pixeltable.com/notebooks/video-generate-ai.md): Create videos from text prompts or animate images using Google's Veo model.
- [Create a video slideshow from images](https://www.pixeltable.com/notebooks/video-image-slideshow.md): Turn a collection of images into an animated video with Ken Burns effects, text overlays, logos, and background music — using a fully declarative Pixeltable pipeline.

## Search & Embeddings

- [Create text embeddings with OpenAI](https://www.pixeltable.com/notebooks/embed-text-openai.md): Generate vector embeddings for text data to enable semantic search and similarity matching.
- [Build semantic search for text](https://www.pixeltable.com/notebooks/search-semantic-text.md): Create a searchable knowledge base that finds content by meaning, not just keywords.
- [Find similar images with CLIP](https://www.pixeltable.com/notebooks/search-similar-images.md): Build visual similarity search to find images that look alike using OpenAI's CLIP model.

## Data Import/Export

- [Import data from CSV files](https://www.pixeltable.com/notebooks/data-import-csv.md): Load data from CSV and Excel files into Pixeltable tables for processing and analysis.
- [Import data from JSON files](https://www.pixeltable.com/notebooks/data-import-json.md): Load structured data from JSON files into Pixeltable tables for processing and analysis.
- [Import data from Parquet files](https://www.pixeltable.com/notebooks/data-import-parquet.md): Load columnar data from Parquet files into Pixeltable tables for processing and analysis.
- [Import data from Excel files](https://www.pixeltable.com/notebooks/data-import-excel.md): Load data from Excel spreadsheets (.xlsx) into Pixeltable tables.
- [Import data from Hugging Face datasets](https://www.pixeltable.com/notebooks/data-import-huggingface.md): Load datasets from Hugging Face Hub into Pixeltable tables for processing with AI models.
- [Load media from S3 and other cloud storage](https://www.pixeltable.com/notebooks/data-import-s3.md): Import images, videos, and audio files from S3, GCS, HTTP URLs, or local paths into Pixeltable tables.
- [Upload media to S3 and other cloud storage](https://www.pixeltable.com/notebooks/data-export-s3.md): When Pixeltable generates media files (thumbnails, extracted frames, processed images), by default it stores them locally.
- [Export data for ML training](https://www.pixeltable.com/notebooks/data-export-pytorch.md): Convert Pixeltable data to PyTorch DataLoader format for model training.
- [Import/Export from SQL Databases](https://www.pixeltable.com/notebooks/data-export-sql.md): Send and receive data from PostgreSQL, SQLite, MySQL, TigerData, or Snowflake for use in external applications.
- [Export data for analytics](https://www.pixeltable.com/notebooks/data-export-analytics.md): Export Pixeltable tables and queries to CSV, JSON, and Apache Iceberg for spreadsheets, scripts, and the lakehouse.
- [Sample data for training and testing](https://www.pixeltable.com/notebooks/data-sampling.md): Create training, validation, and test splits with random or stratified sampling.

## Providers

- [Working with Anthropic in Pixeltable](https://www.pixeltable.com/notebooks/working-with-anthropic.md): Pixeltable's Anthropic integration enables you to access Anthropic's Claude LLM via the Anthropic API.
- [Working with Bedrock in Pixeltable](https://www.pixeltable.com/notebooks/working-with-bedrock.md): Pixeltable's Bedrock integration enables you to access AWS Bedrock foundation models directly from your tables.
- [Working with BFL FLUX in Pixeltable](https://www.pixeltable.com/notebooks/working-with-bfl.md): This notebook demonstrates how to use Black Forest Labs (BFL) FLUX models for image generation and editing through Pixeltable.
- [Working with Deepseek in Pixeltable](https://www.pixeltable.com/notebooks/working-with-deepseek.md): Pixeltable's Deepseek integration enables you to access Deepseek's LLM via the Deepseek API.
- [Working with fal.ai in Pixeltable](https://www.pixeltable.com/notebooks/working-with-fal.md): Pixeltable's fal.ai integration enables you to access fal.ai's fast inference models via the fal.ai API.
- [Working with Microsoft Fabric](https://www.pixeltable.com/notebooks/working-with-fabric.md): Pixeltable's Microsoft Fabric integration enables you to access Azure OpenAI models within Microsoft Fabric notebook environments with automatic authentication.
- [Working with Fireworks AI in Pixeltable](https://www.pixeltable.com/notebooks/working-with-fireworks.md): Pixeltable's Fireworks integration enables you to access LLMs hosted on the Fireworks platform.
- [Working with Gemini in Pixeltable](https://www.pixeltable.com/notebooks/working-with-gemini.md): Use Google Gemini from Pixeltable with a Google AI Studio API key or Vertex AI: generate content, images with Nano Banana and Imagen, and video with Veo.
- [Working with Groq in Pixeltable](https://www.pixeltable.com/notebooks/working-with-groq.md): Pixeltable's Groq integration enables you to access Groq models via the Groq API.
- [Working with Hugging Face](https://www.pixeltable.com/notebooks/working-with-hugging-face.md): Pixeltable provides seamless integration with Hugging Face datasets and models.
- [Working with Jina AI in Pixeltable](https://www.pixeltable.com/notebooks/working-with-jina.md): Pixeltable's Jina AI integration enables you to access state-of-the-art embedding and reranker models via the Jina AI API.
- [Working with llama.cpp in Pixeltable](https://www.pixeltable.com/notebooks/working-with-llama-cpp.md): This tutorial demonstrates how to use Pixeltable's built-in llama.cpp integration to run local LLMs efficiently.
- [Working with Mistral AI in Pixeltable](https://www.pixeltable.com/notebooks/working-with-mistralai.md): Pixeltable's Mistral AI integration enables you to access Mistral's LLM and other models via the Mistral AI API.
- [Working with Nebius](https://www.pixeltable.com/notebooks/working-with-nebius.md): Pixeltable's Nebius Token Factory integration enables you to access Nebius language and embedding models via an OpenAI-compatible API.
- [Working with Ollama in Pixeltable](https://www.pixeltable.com/notebooks/working-with-ollama.md): Ollama is a popular platform for local serving of LLMs. In this tutorial, we'll show how to integrate Ollama models into a Pixeltable workflow.
- [Working with OpenAI in Pixeltable](https://www.pixeltable.com/notebooks/working-with-openai.md): Pixeltable's OpenAI integration enables you to access OpenAI models via the OpenAI API.
- [Working with OpenRouter in Pixeltable](https://www.pixeltable.com/notebooks/working-with-openrouter.md): Pixeltable's OpenRouter integration enables you to access multiple LLM providers through a unified API via OpenRouter.
- [Working with Pydantic in Pixeltable](https://www.pixeltable.com/notebooks/working-with-pydantic.md): Pixeltable's Pydantic integration enables type-safe data insertion using Pydantic models. Instead of inserting raw dictionaries, you can define structured models with validation and insert them directly into Pixeltable tables.
- [Working with Replicate in Pixeltable](https://www.pixeltable.com/notebooks/working-with-replicate.md): Pixeltable's Replicate integration enables you to access Replicate's models via the Replicate API.
- [Working with RunwayML in Pixeltable](https://www.pixeltable.com/notebooks/working-with-runwayml.md): Pixeltable's RunwayML integration lets you generate and transform images and videos using RunwayML's Gen-4.5, Veo, and other models, directly from your Pixeltable tables.
- [Working with Tigris in Pixeltable](https://www.pixeltable.com/notebooks/working-with-tigris.md): This tutorial demonstrates how to configure Pixeltable to use Tigris for storage. This lets you store unlimited amounts of images in Tigris' global data plane, allowing your images to load fast everywhere.
- [Working with Together AI in Pixeltable](https://www.pixeltable.com/notebooks/working-with-together.md): Use Together AI from Pixeltable for chat completions and image generation.
- [Working with Twelve Labs in Pixeltable](https://www.pixeltable.com/notebooks/working-with-twelvelabs.md): Twelve Labs provides multimodal embeddings that project text, images, audio, and video into the same semantic space. This enables true cross-modal search - the most powerful feature of this integration.
- [Working with vLLM](https://www.pixeltable.com/notebooks/working-with-vllm.md): This tutorial demonstrates how to use Pixeltable's built-in vLLM integration to run local LLMs with high-throughput inference.
- [Working with Voyage AI in Pixeltable](https://www.pixeltable.com/notebooks/working-with-voyageai.md): Pixeltable's Voyage AI integration enables you to access state-of-the-art embedding and reranker models via the Voyage AI API.
- [Working with Voxel51 for Visualization in Pixeltable](https://www.pixeltable.com/notebooks/working-with-fiftyone.md): Pixeltable can export data directly from tables and views to the popular Voxel51 frontend, providing a way to visualize and explore image and video datasets.
