Contributing to LlamaIndex#

Interested in contributing to LlamaIndex? Here's how to get started!

QuickStart#

For python users who just want to dive in and start contributing, here's a quick guide on the env setup (if any of this doesn't make sense, read on to the full guide):

Fork the repo and clone your fork
cd llama_index
Setup a new venv with poetry shell
Install dev (and/or docs) dependencies with poetry install --only dev,docs
Install the packages you intend to edit (i.e. pip install -e llama-index-core or pip install -e llama-index-integrations/llms/llama-index-llms-openai)

Contribution Guideline#

The best part of LlamaIndex is our community of users and contributors.

What should I work on?#

🆕 Extend core modules by contributing an integration
📦 Contribute a Tool, Reader, Pack, or Dataset (formerly from llama-hub)
🧠 Add new capabilities to core
🐛 Fix bugs
🎉 Add usage examples
🧪 Add experimental features
📄 Improve code quality & documentation

Also, join our Discord for ideas and discussions: https://discord.gg/dGcwcsnxhU.

1. 🆕 Extend Core Modules#

The most impactful way to contribute to LlamaIndex is by extending our core modules: LlamaIndex modules

We welcome contributions in all modules shown above. So far, we have implemented a core set of functionalities for each, all of which are encapsulated in the LlamaIndex core package. As a contributor, you can help each module unlock its full potential. Provided below are brief description of these modules. You can also refer to their respective folders within this Github repository for some example integrations.

Contributing an integration involves submitting the source code for a new Python package. For now, these integrations will live in the LlamaIndex Github repository and the team will be responsible for publishing the package to PyPi. (Having these packages live outside of this repository and maintained by our community members is in consideration.)

Creating A New Integration Package#

Both llama-index and llama-index-core come equipped with a command-line tool that can be used to initialize a new integration package.

cd ./llama-index-integrations/llms
llamaindex-cli new-package --kind "llms" --name "gemini"

Executing the above commands will create a new folder called llama-index-llms-gemini within the llama-index-integrations/llms directory.

Please ensure to add a detailed README for your new package as it will appear in both llamahub.ai as well as the PyPi.org website. In addition to preparing your source code and supplying a detailed README, we also ask that you fill in some metadata for your package to appear in llamahub.ai with the correct information. You do so by adding the required metadata under the [tool.llamahub] section with your new package's pyproject.toml.

Below is the example of the metadata required for all of our integration packages. Please replace the default author "llama-index" with your own Github user name.

[tool.llamahub]
contains_example = false
import_path = "llama_index.llms.anthropic"

[tool.llamahub.class_authors]
Anthropic = "llama-index"

(source)

Module Details#

Below, we will describe what each module does, give a high-level idea of the interface, show existing implementations, and give some ideas for contribution.

Data Loaders#

A data loader ingests data of any format from anywhere into Document objects, which can then be parsed and indexed.

Interface:

load_data takes arbitrary arguments as input (e.g. path to data), and outputs a sequence of Document objects.
lazy_load_data takes arbitrary arguments as input (e.g. path to data), and outputs an iterable object of Document objects. This is a lazy version of load_data, which is useful for large datasets.

Note: If only lazy_load_data is implemented, load_data will be delegated to it.

Examples:

Contributing a data loader is easy and super impactful for the community.

Ideas

Want to load something but there's no LlamaHub data loader for it yet? Make a PR!

Node Parser#

A node parser parses Document objects into Node objects (atomic units of data that LlamaIndex operates over, e.g., chunk of text, image, or table). It is responsible for splitting text (via text splitters) and explicitly modeling the relationship between units of data (e.g. A is the source of B, C is a chunk after D).

Interface: get_nodes_from_documents takes a sequence of Document objects as input, and outputs a sequence of Node objects.

Examples:

Hierarchical Node Parser

See the API reference for full details.

Ideas:

Add new Node relationships to model hierarchical documents (e.g. play-act-scene, chapter-section-heading).

Text Splitters#

Text splitter splits a long text str into smaller text str chunks with desired size and splitting "strategy" since LLMs have a limited context window size, and the quality of text chunk used as context impacts the quality of query results.

Interface: split_text takes a str as input, and outputs a sequence of str

Examples:

Document/Index/KV Stores#

Under the hood, LlamaIndex also supports a swappable storage layer that allows you to customize Document Stores (where ingested documents (i.e., Node objects) are stored), and Index Stores (where index metadata are stored)

We have an underlying key-value abstraction backing the document/index stores. Currently we support in-memory and MongoDB storage for these stores. Open to contributions!

See the API reference for details.

Managed Index#

A managed index is used to represent an index that's managed via an API, exposing API calls to index documents and query documents.

For example, we support the VectaraIndex. Open to contributions!

See Managed Index docs for details.

Vector Stores#

Our vector store classes store embeddings and support lookup via similarity search. These serve as the main data store and retrieval engine for our vector index.

Interface:

add takes in a sequence of NodeWithEmbeddings and inserts the embeddings (and possibly the node contents & metadata) into the vector store.
delete removes entries given document IDs.
query retrieves top-k most similar entries given a query embedding.
get_nodes get nodes by ID or filters
delete_nodes delete nodes by ID or filters
clear clears an entire db of data

Examples:

Ideas:

See a vector database out there that we don't support yet? A vector store missing methods like get_nodes and delete_nodes? Make a PR!

See reference for full details.

Retrievers#

Our retriever classes are lightweight classes that implement a retrieve method. They may take in an index class as input - by default, each of our indices (list, vector, keyword) has an associated retriever. The output is a set of NodeWithScore objects (a Node object with an extra score field).

You may also choose to implement your own retriever classes on top of your own data if you wish.

Interface:

retrieve takes in a str or QueryBundle as input, and outputs a list of NodeWithScore objects

Examples:

Ideas:

Besides the "default" retrievers built on top of each index, what about fancier retrievers? E.g. retrievers that take in other retrievers as input? Or other types of data?

Query Engines#

Our query engine classes are lightweight classes that implement a query method; the query returns a response type. For instance, they may take in a retriever class as input; our RetrieverQueryEngine takes in a retriever as input as well as a BaseSynthesizer class for response synthesis, and the query method performs retrieval and synthesis before returning the final result. They may take in other query engine classes as input too.

Interface:

query takes in a str or QueryBundle as input, and outputs a Response object.

Examples:

Query Transforms#

A query transform augments a raw query string with associated transformations to improve index querying. This can interpreted as a pre-processing stage, before the core index query logic is executed.

Interface: run takes in a str or Querybundle as input, and outputs a transformed QueryBundle.

Examples:

See guide for more information.

Node Postprocessors#

A node postprocessor refines a list of retrieved nodes given configuration and context.

Interface: postprocess_nodes takes a list of Nodes and extra metadata (e.g. similarity and query), and outputs a refined list of Nodes.

Examples:

Keyword Postprocessor: filters nodes based on keyword match
Colbert Rerank Postprocessor: reranks retrieved nodes.
Presidio Postprocessor: provides some data privacy on retrieved nodes by omitting personal information.

Output Parsers#

An output parser enables us to extract structured output from the plain text output generated by the LLM.

Interface:

format: formats a query str with structured output formatting instructions, and outputs the formatted str
parse: takes a str (from LLM response) as input, and gives a parsed structured output (optionally also validated, error-corrected).

Examples:

See guide for more information.

2. 📦 Contribute a Pack, Reader, Tool, or Dataset (formerly from llama-hub)#

Contributing a new Reader or Tool involves submitting a new package within the llama-index-integrations/readers and llama-index-integrations/tools, folders respectively.

The LlamaIndex command-line tool can be used to initialize new Packs and Integrations. (NOTE: llama-index-cli comes installed with llama-index.)

cd ./llama-index-packs
llamaindex-cli new-package --kind "packs" --name "my new pack"

cd ./llama-index-integrations/readers
llamaindex-cli new-package --kind "readers" --name "new reader"

Executing the first set of shell commands will create a new folder called llama-index-packs-my-new-pack within the llama-index-packs directory. While the second set will create a new package directory called llama-index-readers-new-reader within the llama-index-integrations/readers directory.

Please ensure to add a detailed README for your new package as it will appear in both llamahub.ai as well as the PyPi.org website. In addition to preparing your source code and supplying a detailed README, we also ask that you fill in some metadata for your package to appear in llamahub.ai with the correct information. You do so by adding the required metadata under the [tool.llamahub] section with your new package's pyproject.toml.

Below is the example of the metadata required for packs, readers and tools:

[tool.llamahub]
contains_example = true
import_path = "llama_index.packs.agent_search_retriever"

[tool.llamahub.class_authors]
AgentSearchRetrieverPack = "logan-markewich"

(source)

3. 🧠 Add new capabilities to core#

We would greatly appreciate any and all contributions to our core abstractions that represent enhancements from the current set of capabilities. General improvements that make these core abstractions more robust and thus easier to build on are also welcome!

4. 🐛 Fix Bugs#

Most bugs are reported and tracked in the Github Issues Page. We try our best in triaging and tagging these issues:

Issues tagged as bug are confirmed bugs.
New contributors may want to start with issues tagged with good first issue.

Please feel free to open an issue and/or assign an issue to yourself.

5. 🎉 Add Usage Examples#

If you have applied LlamaIndex to a unique use-case (e.g. interesting dataset, customized index structure, complex query), we would love your contribution in the form of:

a guide: e.g. Guide to LlamIndex + Structured Data
an example notebook: e.g. Email Info Extraction

6. 🧪 Add Experimental Features#

If you have a crazy idea, make a PR for it! Whether if it's the latest research, or what you thought of in the shower, we'd love to see creative ways to improve LlamaIndex.

7. 📄 Improve Code Quality & Documentation#

We would love your help in making the project cleaner, more robust, and more understandable. If you find something confusing, it most likely is for other people as well. Help us be better!

Development Guidelines#

Repo Structure#

The llama_index repo is structured as a mono-repo of many packages. For example, llama-index-core/, llama-index-integrations/llms/llama-index-llms-openai, and llama-index-integrations/embeddings/llama-index-embeddings-openai are all separate python packages. This organization should hopefully direct you to where you want to make a change or add a new modules.

Setting up environment#

LlamaIndex is a Python package. We've tested primarily with Python versions >= 3.8. Here's a quick and dirty guide to setting up your environment for local development.

Fork LlamaIndex Github repo* and clone it locally. (New to GitHub / git? Here's how.)
In a terminal, cd into the directory of your local clone of your forked repo.
Install pre-commit hooks* by running pre-commit install. These hooks are small house-keeping scripts executed every time you make a git commit, which automates away a lot of chores.
Prepare a virtual environment.
Install Poetry*. This will help you manage package dependencies.
Execute poetry shell. This command will create a virtual environment specific for this package, which keeps installed packages contained to this project. (New to Poetry, the dependency & packaging manager for Python? Read about its basic usage here.)
Execute poetry install --only dev,docs*. This will install all dependencies needed for local development. To see what will be installed, read the pyproject.toml under that directory.
cd into the specific package you want to work on. For example, if I want to work on the core package, I execute cd llama-index-core/. (New to terminal / command line? Here's a getting started guide.)
Install that specific integration with pip install -e . (or alternatively, pip install -e <path to package>). This will install the package in editable mode, which means any changes you make to that package will show up when you run your code again. NOTE: If working in a notebook, you will need to restart it for changes to packages to show up.

Steps marked with an asterisk (*) are one-time tasks. You don't have to repeat them when you attempt to contribute on something else next time.

Now you should be set!

Validating your Change#

Let's make sure to format/lint our change. For bigger changes, let's also make sure to test it and perhaps create an example notebook.

Formatting/Linting#

We run an assortment of linters: black, ruff, mypy.

If you have installed pre-commit hooks in this repo, they should have taken care of the formatting and linting automatically.

If -- for whatever reason -- you would like to do it manually, you can format and lint your changes with the following commands in the root directory:

make format; make lint

Under the hood, we still install pre-commit hooks for you, so that you don't have to do this manually next time.

Testing#

If you modified or added code logic, create test(s), because they help preventing other maintainers from accidentally breaking the nice things you added / re-introducing the bugs you fixed.

In almost all cases, add unit tests.
If your change involves adding a new integration, also add integration tests. When doing so, please mock away the remote system that you're integrating LlamaIndex with, so that when the remote system changes, LlamaIndex developers won't see test failures.

Reciprocally, you should run existing tests (from every package that you touched) before making a git commit, so that you can be sure you didn't break someone else's good work.

(By the way, when a test is run with the goal of detecting whether something broke in a new version of the codebase, it's referred to as a "regression test". You'll also hear people say "the test regressed" as a more diplomatic way of saying "the test failed".)

Our tests are stored in the tests folders under each package directory. We use the testing framework pytest, so you can just run pytest in each package you touched to run all its tests.

Regardless of whether you have run them locally, a CI system will run all affected tests on your PR when you submit one anyway. There, tests are orchestrated with Pants, the build system of our choice. There is a slight chance that tests broke on CI didn't break on your local machine or the other way around. When that happens, please take our CI as the source of truth. This is because our release pipeline (which builds the packages users are going to download from PyPI) are run in the CI, not on your machine (even if you volunteer), so it's the CI that is the golden standard.

Creating an Example Notebook#

For changes that involve entirely new features, it may be worth adding an example Jupyter notebook to showcase this feature.

Example notebooks can be found in this folder.

Creating a pull request#

See these instructions to open a pull request against the main LlamaIndex repo.