Metadata-Version: 2.1
Name: gptstonks-wrappers
Version: 0.0.3
Summary: Useful wrappers around common AI tools: LangChain, LlamaIndex, etc.
Author-Email: GPTStonks Team <gptstonks@gmail.com>
License: MIT
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Project-URL: Homepage, https://github.com/GPTStonks/gptstonks
Project-URL: Bug Tracker, https://github.com/GPTStonks/gptstonks/issues
Requires-Python: <3.12,>=3.10
Requires-Dist: rank-bm25>=0.2.2
Requires-Dist: llama-index>=0.10.13
Requires-Dist: langchain<0.4,>=0.3
Requires-Dist: llama-index-llms-openai>=0.1.6
Requires-Dist: llama-index-retrievers-bm25>=0.1.3
Requires-Dist: llama-index-readers-file>=0.1.33
Requires-Dist: langchain-community<0.4,>=0.3
Provides-Extra: huggingface
Requires-Dist: llama-index-embeddings-huggingface>=0.1.3; extra == "huggingface"
Requires-Dist: llama-index-llms-huggingface>=0.1.3; extra == "huggingface"
Provides-Extra: testing
Requires-Dist: pytest; extra == "testing"
Requires-Dist: pytest-cov; extra == "testing"
Requires-Dist: trio>=0.23.2; extra == "testing"
Requires-Dist: sentence-transformers>=2.2.2; extra == "testing"
Requires-Dist: pytest-asyncio>=0.23.5; extra == "testing"
Requires-Dist: duckduckgo-search>=4.5.0; extra == "testing"
Requires-Dist: wikipedia>=1.4.0; extra == "testing"
Requires-Dist: pinecone-client>=3.2.2; extra == "testing"
Requires-Dist: llama-index-vector-stores-pinecone>=0.1.6; extra == "testing"
Description-Content-Type: text/markdown

<p align="center">
  <img src="../../docs/assets/logo-chatbot.png" alt="Logo">
</p>
<p align="center">
  <!-- Waitlist Badge -->
  <a href="https://gptstonks.net/login"><img src="../../docs/assets/waitlist_badge.png" alt="Join Waitlist Badge"></a>
  <!-- YT Badge -->
  <a href="https://www.youtube.com/@GPTStonks"><img src="https://img.shields.io/badge/channel-ff0000?style=for-the-badge&logo=youtube&logoColor=white" alt="Youtube Channel Badge"></a>
  <!-- X Badge -->
  <a href="https://twitter.com/GPTStonks"><img src="https://img.shields.io/badge/follow_us-000000?style=for-the-badge&logo=x&logoColor=white" alt="X Follow Us Badge"></a>
  <!-- Discord Badge -->
  <a href="https://discord.gg/MyDDGuEd"><img src="https://img.shields.io/badge/Discord-5865F2?style=for-the-badge&logo=discord&logoColor=white" alt="Discord Badge"></a>
  <!-- Docker Badge -->
  <a href="https://hub.docker.com/u/gptstonks">
    <img src="https://img.shields.io/badge/Docker-2496ED?style=for-the-badge&logo=docker&logoColor=white" alt="Docker Badge">
  </a>
</p>
<p align="center">
  <!-- Hugging Face Badge -->
  <a href="https://huggingface.co/"><img src="https://img.shields.io/badge/Hugging%20Face-F58025?style=for-the-badge&logo=huggingface&logoColor=white" alt="Hugging Face Badge"></a>
  <!-- LangChain Badge -->
  <a href="https://langchain.com/">
    <img src="https://img.shields.io/badge/LangChain-005A9C?style=for-the-badge&logo=langchain&logoColor=white" alt="LangChain Badge">
  </a>
  <!-- LlamaIndex Badge -->
  <a href="https://www.llamaindex.ai/"><img src="https://img.shields.io/badge/LlamaIndex-ac6aff?style=for-the-badge&logo=llamaindex&logoColor=white" alt="LlamaIndex Badge"></a>
</p>

# GPTStonks Wrappers

## Description

GPTStonks Wrappers provides Auto models, similar to the `transformers` library, but for common AI tools instead of models: LangChain, LlamaIndex, etc.

## Development

  1. Install [PDM](https://pdm.fming.dev/latest/#installation).

  2. Clone the project and install necessary packages:
```bash
# clone project
git clone https://github.com/GPTStonks/gptstonks.git
cd gptstonks

# install pdm
pip install pdm

# install package
pdm install -dG default
```

## Sample usage with pre-trained models

In the [API project](../../projects/gptstonks_api/), `AutoRag` is used to perform [retrieval-augmented generation](https://arxiv.org/abs/2005.11401) (RAG) with [OpenBB](https://openbb.co)'s official documentation and with pre-trained models (e.g., OpenAI, Anthropic, Llama.cpp, etc.). Additionally, `AutoMultiStepQueryEngine` plans and executes Internet searches to solve complex queries.
