indexify

A realtime serving engine for Data-Intensive Generative AI Applications

Stars: 980

Visit

Indexify is an open-source engine for building fast data pipelines for unstructured data (video, audio, images, and documents) using reusable extractors for embedding, transformation, and feature extraction. LLM Applications can query transformed content friendly to LLMs by semantic search and SQL queries. Indexify keeps vector databases and structured databases (PostgreSQL) updated by automatically invoking the pipelines as new data is ingested into the system from external data sources. **Why use Indexify** * Makes Unstructured Data **Queryable** with **SQL** and **Semantic Search** * **Real-Time** Extraction Engine to keep indexes **automatically** updated as new data is ingested. * Create **Extraction Graph** to describe **data transformation** and extraction of **embedding** and **structured extraction**. * **Incremental Extraction** and **Selective Deletion** when content is deleted or updated. * **Extractor SDK** allows adding new extraction capabilities, and many readily available extractors for **PDF**, **Image**, and **Video** indexing and extraction. * Works with **any LLM Framework** including **Langchain**, **DSPy**, etc. * Runs on your laptop during **prototyping** and also scales to **1000s of machines** on the cloud. * Works with many **Blob Stores**, **Vector Stores**, and **Structured Databases** * We have even **Open Sourced Automation** to deploy to Kubernetes in production.

README:

Indexify

Create and Deploy Durable, Data-Intensive Agentic Workflows

Indexify simplifies building and serving durable, multi-stage data-intensive workflows and exposes them as HTTP APIs or Python Remote APIs.

[!NOTE]
Indexify is the Open-Source core compute engine that powers Tensorlake's Serverless Workflow Engine for processing unstructured data.

💡 Use Cases

Indexify is a versatile data processing framework for all kinds of use cases, including:

⭐ Key Features

Multi-Cloud/Datacenter/Region: Leverage Compute in your workflows from other clouds with very little hassle and configuration.
Distributed Processing: Run functions in parallel across machines for scaleouts use-cases.
Resource Allocation: Span workflows across GPU and CPU instances so that functions can be assigned to their optimal hardware.
Dynamic Routing: Route data to different specialized compute functions distributed on a cluster based on conditional branching logic.

⚙️ Installation

Install the Tensorlake SDK for building workflows and the Indexify CLI.

pip install indexify tensorlake

📚 A Minimal Example

Functions decorated with @tensorlake_function() are units of compute in your Workflow APIs. These functions can have data dependencies on other functions.
Tensorlake functions are durable, i.e, if the function crashes or the node running the function is lost, it is going to be automatically retried on another running instance with the same input.
You can run as many function instances on the cluster, inputs are going to be automatically load balanced across them when the workflow is called in parallel by other applications.

The example below is a workflow API that accepts some text, embeds the file, and writes it to a local vector-db. Each function could be placed on different classes of machines(CPU-only machines for chunking and writing to databases, NVIDIA GPUs for embedding)

from pydantic import BaseModel
from tensorlake import tensorlake_function, Graph, Image, TensorlakeCompute
from typing import List, Union

# Define Input and Outputs of various functions in your workflow
class Text(BaseModel):
    text: str


class TextChunk(BaseModel):
    chunk: str
    page_number: int


class ChunkEmbedding(BaseModel):
    text: str
    embedding: List[float]

# Define an image capable of running the functions. Each image
# can have their own image
embedding_image = (
    Image()
    .name("text_embedding_image")
    .run("pip install langchain")
    .run("pip install sentence_transformer")
    .run("pip install langchain-text-splitters")
    .run("pip install chromadb")
    .run("pip install uuid")
)


# Chunk the text for embedding and retrieval
@tensorlake_function(input_encoder="json", image=embedding_image)
def chunk_text(input: dict) -> List[TextChunk]:
    text = Text.model_validate(input)
    from langchain_text_splitters import RecursiveCharacterTextSplitter

    text_splitter = RecursiveCharacterTextSplitter(
        chunk_size=1000,
        chunk_overlap=20,
        length_function=len,
        is_separator_regex=False,
    )
    texts = text_splitter.create_documents([text.text])
    return [
        TextChunk(chunk=chunk.page_content, page_number=i)
        for i, chunk in enumerate(texts)
    ]


# Embed a single chunk.
# Note: (Automatic Map) Indexify automatically parallelize functions when they consume an
# element from functions that produces a List. In this case each text chunk is processed
# in parallel by an Embedder function
class Embedder(TensorlakeCompute):
    name = "embedder"
    image = embedding_image

    # TensorlakeCompute function allows initializing resources in the constructors
    # and they are not unloaded again until the compute object is destroyed.  
    def __init__(self):
        from sentence_transformers import SentenceTransformer
        self._model = SentenceTransformer("all-MiniLM-L6-v2")

    def run(self, chunk: TextChunk) -> ChunkEmbedding:
        embeddings = self._model.encode(chunk.chunk)
        return ChunkEmbedding(text=chunk.chunk, embedding=embeddings)


class EmbeddingWriter(TensorlakeCompute):
    name = "embedding_writer"
    image = embedding_image

    def __init__(self):
        import chromadb

        self._chroma = chromadb.PersistentClient("./chromadb_tensorlake")
        self._collection = collection = self._chroma.create_collection(
            name="my_collection", get_or_create=True
        )

    def run(self, embedding: ChunkEmbedding) -> None:
        import uuid

        self._collection.upsert(
            ids=[str(uuid.uuid4())],
            embeddings=[embedding.embedding],
            documents=[embedding.text],
        )


# Constructs a compute graph connecting the three functions defined above into a workflow that generates
# runs them as a pipeline
graph = Graph(
    name="text_embedder",
    start_node=chunk_text,
    description="Splits, embeds and indexes text",
)
graph.add_edge(chunk_text, Embedder)
graph.add_edge(Embedder, EmbeddingWriter)

Testing Locally

You can test the workflow locally with only the tensorlake package installed.

invocation_id = graph.run(input={"text": "This is a test text"})
print(f"Invocation ID: {invocation_id}")

# You can get output from each function of the graph
embedding = graph.output(invocation_id, "embedder")
print(embedding)

Deploying Workflows as APIs

Big Picture, you will deploy Indexify Server on a machine and run containers for each of the function in a workflow separately.

But first, we will show how to do this locally on a single machine.

Start the Server

Download a server release from here. Open a terminal and start the server.

./indexify-server -dev

Start the Executors

Executor is the component which is responsible for running your functions. On a terminal, where all the dependencies are installed, start an executor in development mode.

indexify-cli executor --dev

Set the environment variable -

export INDEXIFY_URL=http://localhost:8900

Change the code in the workflow to the following -

from tensorlake import RemoteGraph
RemoteGraph.deploy(graph)

At this point, you now have a Graph endpoint on Indexify Server ready to be called as an API from any application.

Invoke the Graph

You can invoke the Graph as a REST API if the first function is configured to accept JSON payload.

curl -X 'POST' http://localhost:8900/namespaces/default/compute_graphs/text_embedder/invoke_object -H 'Content-Type: application/json' -d '{"input": {"text": "hello world"}}'

This returns you an invocation id - {"id":"55df51b4a84ffc69"}. An Invocation Id can be used to get the status of the workflow as it processes that input, and getting any outputs off the graph.

Get the outputs of the Embedding function -

curl -X GET http://localhost:8900/namespaces/default/compute_graphs/text_embedder/invocations/55df51b4a84ffc69/outputs

This returns all the outputs of the function -

{"status":"finalized","outputs":[{"compute_fn":"chunk_text","id":"89de2063abadf5d3","created_at":1738110077424},{"compute_fn":"embedder","id":"4908f00d711c4cd1","created_at":1738110081015}],"cursor":null}

You can now retrieve one of the outputs -

curl -X GET http://localhost:8900/namespaces/default/compute_graphs/text_embedder/invocations/55df51b4a84ffc69/fn/embedder/output/4908f00d711c4cd1

You can invoke the Graph from Python too

from tensorlake import RemoteGraph
remote_graph = RemoteGraph.by_name("text_embedder")

Deploying to Production Clusters

Deploying a workflow to production is a two step process -

1. Run the server

docker run -it -p 8900:8900 tensorlake/indexify-server

2. Building and Deploying Function Containers

First build and deploy container images that contains all the python and system dependencies of your code. They can be built using standard Docker build systems. For this example, we have a single image that can run all the functions. You can separate them to reduce the size of your images for more complex projects.

indexify-cli build-image workflow.py

This builds the following image, as defined in the workflow code above - text_embedding_image

Next Deploy the Containers

docker run --it text_embedding_image indexify-cli executor --function default:text_embedder:chunk_document
docker run --it text_embedding_image indexify-cli executor --function default:text_embedder:embed_chunk
docker run --it text_embedding_image indexify-cli executor --function default:text_embedder:write_to_db

Containers are treated as ephemeral, only one type of function is ever scheduled on a single container. We are starting two containers for placing one function in each of them.

Scale To Zero

Indexify won't complain if you shut down the containers at night. It will still accept new API calls from external systems even it can't find machines to run functions. It will simply queue them up, and wait for functions to come up. It emits telemetry of pending tasks, waiting to be placed on functions which can be used as inputs to Autoscalers.

This is it!

Production Ready Distributed and Always-On Data Workflows

You have built a workflow API which is durable, capable of being distributed on many kinds of hardware and can handle limitless scale.
You can use any Python libraries under the sun, any system packages and can use your favorite tools to package them into a container image.
Deploying code is as simple as uploading code into the server, they get distributed and updated automatically.

🗺️ Roadmap

⏳ Scheduler

Function Batching: Process multiple functions in a single batch to improve efficiency.
Data Localized Execution: Boost performance by prioritizing execution on machines where intermediate outputs exist already.
Reducer Optimizations: Optimize performance by batching the serial execution of reduced function calls.
Parallel Scheduling: Reduce latency by enabling parallel execution across multiple machines.
Cyclic Graph Support: Enable more flexible agentic behaviors by leveraging cycles in graphs.
Ephemeral Graphs: Perform multi-stage inference and retrieval without persisting intermediate outputs.
Data Loader Functions: Stream values into graphs over time using the yield keyword.

🛠️ SDK

TypeScript SDK: Build an SDK for writing workflows in Typescript.

For Tasks:

Click tags to check more tools for each tasks

extract data from videos extract data from audio extract data from images extract data from documents query data with sql

For Jobs:

data engineer data scientist machine learning engineer research scientist software engineer

Alternative AI tools for indexify

Similar Open Source Tools

indexify

github

: 980

open-parse

Open Parse is a Python library for visually discerning document layouts and chunking them effectively. It is designed to fill the gap in open-source libraries for handling complex documents. Unlike text splitting, which converts a file to raw text and slices it up, Open Parse visually analyzes documents for superior LLM input. It also supports basic markdown for parsing headings, bold, and italics, and has high-precision table support, extracting tables into clean Markdown formats with accuracy that surpasses traditional tools. Open Parse is extensible, allowing users to easily implement their own post-processing steps. It is also intuitive, with great editor support and completion everywhere, making it easy to use and learn.

github

: 2.4k

restai

RestAI is an AIaaS (AI as a Service) platform that allows users to create and consume AI agents (projects) using a simple REST API. It supports various types of agents, including RAG (Retrieval-Augmented Generation), RAGSQL (RAG for SQL), inference, vision, and router. RestAI features automatic VRAM management, support for any public LLM supported by LlamaIndex or any local LLM supported by Ollama, a user-friendly API with Swagger documentation, and a frontend for easy access. It also provides evaluation capabilities for RAG agents using deepeval.

github

: 414

superduper

superduper.io is a Python framework that integrates AI models, APIs, and vector search engines directly with existing databases. It allows hosting of models, streaming inference, and scalable model training/fine-tuning. Key features include integration of AI with data infrastructure, inference via change-data-capture, scalable model training, model chaining, simple Python interface, Python-first approach, working with difficult data types, feature storing, and vector search capabilities. The tool enables users to turn their existing databases into centralized repositories for managing AI model inputs and outputs, as well as conducting vector searches without the need for specialized databases.

github

: 5.0k

postgresml

PostgresML is a powerful Postgres extension that seamlessly combines data storage and machine learning inference within your database. It enables running machine learning and AI operations directly within PostgreSQL, leveraging GPU acceleration for faster computations, integrating state-of-the-art large language models, providing built-in functions for text processing, enabling efficient similarity search, offering diverse ML algorithms, ensuring high performance, scalability, and security, supporting a wide range of NLP tasks, and seamlessly integrating with existing PostgreSQL tools and client libraries.

github

: 6.1k

llm-answer-engine

This repository contains the code and instructions needed to build a sophisticated answer engine that leverages the capabilities of Groq, Mistral AI's Mixtral, Langchain.JS, Brave Search, Serper API, and OpenAI. Designed to efficiently return sources, answers, images, videos, and follow-up questions based on user queries, this project is an ideal starting point for developers interested in natural language processing and search technologies.

github

: 4.5k

langmanus

LangManus is a community-driven AI automation framework that combines language models with specialized tools for tasks like web search, crawling, and Python code execution. It implements a hierarchical multi-agent system with agents like Coordinator, Planner, Supervisor, Researcher, Coder, Browser, and Reporter. The framework supports LLM integration, search and retrieval tools, Python integration, workflow management, and visualization. LangManus aims to give back to the open-source community and welcomes contributions in various forms.

github

: 2.9k

kollektiv

Kollektiv is a Retrieval-Augmented Generation (RAG) system designed to enable users to chat with their favorite documentation easily. It aims to provide LLMs with access to the most up-to-date knowledge, reducing inaccuracies and improving productivity. The system utilizes intelligent web crawling, advanced document processing, vector search, multi-query expansion, smart re-ranking, AI-powered responses, and dynamic system prompts. The technical stack includes Python/FastAPI for backend, Supabase, ChromaDB, and Redis for storage, OpenAI and Anthropic Claude 3.5 Sonnet for AI/ML, and Chainlit for UI. Kollektiv is licensed under a modified version of the Apache License 2.0, allowing free use for non-commercial purposes.

github

: 74

clearml-server

ClearML Server is a backend service infrastructure for ClearML, facilitating collaboration and experiment management. It includes a web app, RESTful API, and file server for storing images and models. Users can deploy ClearML Server using Docker, AWS EC2 AMI, or Kubernetes. The system design supports single IP or sub-domain configurations with specific open ports. ClearML-Agent Services container allows launching long-lasting jobs and various use cases like auto-scaler service, controllers, optimizer, and applications. Advanced functionality includes web login authentication and non-responsive experiments watchdog. Upgrading ClearML Server involves stopping containers, backing up data, downloading the latest docker-compose.yml file, configuring ClearML-Agent Services, and spinning up docker containers. Community support is available through ClearML FAQ, Stack Overflow, GitHub issues, and email contact.

github

: 364

aistore

AIStore is a lightweight object storage system designed for AI applications. It is highly scalable, reliable, and easy to use. AIStore can be deployed on any commodity hardware, and it can be used to store and manage large datasets for deep learning and other AI applications.

github

: 1.4k

sd-webui-agent-scheduler

AgentScheduler is an Automatic/Vladmandic Stable Diffusion Web UI extension designed to enhance image generation workflows. It allows users to enqueue prompts, settings, and controlnets, manage queued tasks, prioritize, pause, resume, and delete tasks, view generation results, and more. The extension offers hidden features like queuing checkpoints, editing queued tasks, and custom checkpoint selection. Users can access the functionality through HTTP APIs and API callbacks. Troubleshooting steps are provided for common errors. The extension is compatible with latest versions of A1111 and Vladmandic. It is licensed under Apache License 2.0.

github

: 646

inngest

Inngest is a platform that offers durable functions to replace queues, state management, and scheduling for developers. It allows writing reliable step functions faster without dealing with infrastructure. Developers can create durable functions using various language SDKs, run a local development server, deploy functions to their infrastructure, sync functions with the Inngest Platform, and securely trigger functions via HTTPS. Inngest Functions support retrying, scheduling, and coordinating operations through triggers, flow control, and steps, enabling developers to build reliable workflows with robust support for various operations.

github

: 2.6k

UFO

UFO is a UI-focused dual-agent framework to fulfill user requests on Windows OS by seamlessly navigating and operating within individual or spanning multiple applications.

github

: 6.6k

llama-cpp-agent

The llama-cpp-agent framework is a tool designed for easy interaction with Large Language Models (LLMs). Allowing users to chat with LLM models, execute structured function calls and get structured output (objects). It provides a simple yet robust interface and supports llama-cpp-python and OpenAI endpoints with GBNF grammar support (like the llama-cpp-python server) and the llama.cpp backend server. It works by generating a formal GGML-BNF grammar of the user defined structures and functions, which is then used by llama.cpp to generate text valid to that grammar. In contrast to most GBNF grammar generators it also supports nested objects, dictionaries, enums and lists of them.

github

: 454

sec-parser

The `sec-parser` project simplifies extracting meaningful information from SEC EDGAR HTML documents by organizing them into semantic elements and a tree structure. It helps in parsing SEC filings for financial and regulatory analysis, analytics and data science, AI and machine learning, causal AI, and large language models. The tool is especially beneficial for AI, ML, and LLM applications by streamlining data pre-processing and feature extraction.

github

: 99

fastagency

FastAgency is a powerful tool that leverages the AutoGen framework to quickly build applications with multi-agent workflows. It supports various interfaces like ConsoleUI and MesopUI, allowing users to create interactive applications. The tool enables defining workflows between agents, such as students and teachers, and summarizing conversations. FastAgency aims to expand its capabilities by integrating with additional agentic frameworks like CrewAI, providing more options for workflow definition and AI tool integration.

github

: 341

For similar tasks

indexify

github

: 980

unstructured

The `unstructured` library provides open-source components for ingesting and pre-processing images and text documents, such as PDFs, HTML, Word docs, and many more. The use cases of `unstructured` revolve around streamlining and optimizing the data processing workflow for LLMs. `unstructured` modular functions and connectors form a cohesive system that simplifies data ingestion and pre-processing, making it adaptable to different platforms and efficient in transforming unstructured data into structured outputs.

github

: 10.5k

For similar jobs

db2rest

DB2Rest is a modern low-code REST DATA API platform that simplifies the development of intelligent applications. It seamlessly integrates existing and new databases with language models (LMs/LLMs) and vector stores, enabling the rapid delivery of context-aware, reasoning applications without vendor lock-in.

github

: 167

mage-ai

Mage is an open-source data pipeline tool for transforming and integrating data. It offers an easy developer experience, engineering best practices built-in, and data as a first-class citizen. Mage makes it easy to build, preview, and launch data pipelines, and provides observability and scaling capabilities. It supports data integrations, streaming pipelines, and dbt integration.

github

: 7.8k

airbyte

Airbyte is an open-source data integration platform that makes it easy to move data from any source to any destination. With Airbyte, you can build and manage data pipelines without writing any code. Airbyte provides a library of pre-built connectors that make it easy to connect to popular data sources and destinations. You can also create your own connectors using Airbyte's no-code Connector Builder or low-code CDK. Airbyte is used by data engineers and analysts at companies of all sizes to build and manage their data pipelines.

github

: 17.6k

labelbox-python

Labelbox is a data-centric AI platform for enterprises to develop, optimize, and use AI to solve problems and power new products and services. Enterprises use Labelbox to curate data, generate high-quality human feedback data for computer vision and LLMs, evaluate model performance, and automate tasks by combining AI and human-centric workflows. The academic & research community uses Labelbox for cutting-edge AI research.

github

: 134

telemetry-airflow

This repository codifies the Airflow cluster that is deployed at workflow.telemetry.mozilla.org (behind SSO) and commonly referred to as "WTMO" or simply "Airflow". Some links relevant to users and developers of WTMO: * The `dags` directory in this repository contains some custom DAG definitions * Many of the DAGs registered with WTMO don't live in this repository, but are instead generated from ETL task definitions in bigquery-etl * The Data SRE team maintains a WTMO Developer Guide (behind SSO)

github

: 185

airflow

Apache Airflow (or simply Airflow) is a platform to programmatically author, schedule, and monitor workflows. When workflows are defined as code, they become more maintainable, versionable, testable, and collaborative. Use Airflow to author workflows as directed acyclic graphs (DAGs) of tasks. The Airflow scheduler executes your tasks on an array of workers while following the specified dependencies. Rich command line utilities make performing complex surgeries on DAGs a snap. The rich user interface makes it easy to visualize pipelines running in production, monitor progress, and troubleshoot issues when needed.

github

: 39.2k

airbyte-platform

Airbyte is an open-source data integration platform that makes it easy to move data from any source to any destination. With Airbyte, you can build and manage data pipelines without writing any code. Airbyte provides a library of pre-built connectors that make it easy to connect to popular data sources and destinations. You can also create your own connectors using Airbyte's low-code Connector Development Kit (CDK). Airbyte is used by data engineers and analysts at companies of all sizes to move data for a variety of purposes, including data warehousing, data analysis, and machine learning.

github

: 229

chronon

Chronon is a platform that simplifies and improves ML workflows by providing a central place to define features, ensuring point-in-time correctness for backfills, simplifying orchestration for batch and streaming pipelines, offering easy endpoints for feature fetching, and guaranteeing and measuring consistency. It offers benefits over other approaches by enabling the use of a broad set of data for training, handling large aggregations and other computationally intensive transformations, and abstracting away the infrastructure complexity of data plumbing.

github

: 766