The development of AI cannot be separated from the contribution of the open source community. It is precisely because of the dedicated efforts of outstanding engineers worldwide that AI is now in full swing

The development of AI cannot be separated from the contribution of the open source community. It is precisely because of the dedicated efforts of outstanding engineers worldwide that AI is now in full swing. Today, the editor has compiled some popular open source projects related to AI for enthusiasts to learn and research.

Training model

DeepSpeed Microsoft

DeepSpeed is a ChatGPT model launched by Microsoft for individuals, making distributed training and reasoning simple, efficient, and effective. Rumor has it that the performance is fast and the cost is low.

As is well known, due to the lack of OpenAI, the open source community has launched models such as LLaMa, Alpaca, Vicuna, and Databricks Dolly to make ChatGPT like models more accessible to people.

However, due to the lack of an end-to-end RLHF scaling system, the training of ChatGPT like models is still very difficult. The emergence of DeepSpeedChat has just completed this issue.

Address: https://github.com/microsoft/DeepSpeed

JARVIS Microsoft

Microsoft has launched a private smart butler system that connects LLM with the ML community. The same is also to solve the problem of AI model connectivity. Paper: https://arxiv.org/pdf/2303.17580.pdf

Address: https://github.com/microsoft/JARVIS

Programming tools

LLAMA OpenAIAPIAdapter

Llama.cpp is packaged as OpenAIAPI, which provides a model quantification and deployment method on local CPUs.

Reference paper 'LLaMA AdapterV2: Parameter Efficient Visual Instruction Model'

Address: https://github.com/abetlen/llama-cpp-python

LangChain

It is a powerful framework designed to help developers build end-to-end applications using language models. It provides a set of tools, components, and interfaces to simplify the process of creating applications supported by large language models (LLM) and chat models. LangChain can easily manage interactions with language models, link multiple components together, and integrate additional resources such as APIs and databases.

Address: https://langchain.com

AutoGPT

As the name suggests, when you give AutoGPT a question, it will think for itself, provide implementation steps and details, which is a milestone development for AI.

However, it requires local deployment and is relatively expensive due to token usage.

Address: https://github.com/Torantulino/Auto-GPT

LangFlow

LangFlow is a native LLM graphical development interface based on LangChain. If you are familiar with LangChain, the development interface for Chains, Agents, and PromptEngineering will feel very intuitive. A workflow system built based on LangChain and React Flow. Once the workflow is generated, it can be exported to Json format, and then LangChain can directly call it

Address: http://github.com/logspace-ai/langflow

CoGnoSYS

Tools similar to AutoGPT, the online version of AutoGPT is better to use, has an interface, and does not require local deployment.

Address: http://cognosys.ai

AI Tower - Aridama Hall

The various online models launched by Alibaba Dharma Academy can be jointly debugged, making them more user-friendly for domestic users.

Address: www.modelscope.cn

Image Tools

SegmentAnything

SegmentAnything

Address: https://github.com/IDEA-Research/Grounded-Segment-Anything

MagicCopy

Chrome extension tool, image Object detection matting. Can quickly process images.

Address: https://github.com/kevmo314/magic-copy

PlaskPost

Pose image generator, 3D model+image generation for specified actions

Address: https://plask.ai/

VideoCrafter

One click replacement, text to video support for open source editing functions.

Address: https://github.com/VideoCrafter/VideoCrafter

Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])