Baichuan-13B

  • PROJECT DIRECT URL: https://github.com/baichuan-inc/Baichuan-13B
  • DESCRIPTION: Baichuan-13B is a large language model with 13 billion parameters, developed by Baichuan Intelligent Technology. It’s optimized for both Chinese and English processing, following the Transformer architecture, and aimed at providing a robust tool for a wide range of NLP tasks. This model emphasizes accessibility and commercial application, making significant strides in AI and machine learning, especially for applications requiring bilingual capabilities​​​​​.

OpenLLMs

  • PROJECT DIRECT URL: https://github.com/eugeneyan/open-llms
  • DESCRIPTION: OpenLLMs is a curated list of open-source large language models (LLMs) available for commercial use. This project aims to provide researchers and developers with easy access to a wide range of LLMs for various applications, facilitating experimentation and development in the field of natural language processing and artificial intelligence.

OpenChat

  • PROJECT DIRECT URL: https://github.com/imoneoi/openchat
  • DESCRIPTION: OpenChat focuses on advancing open-source language models by addressing challenges with imperfect data. It showcases a range of models from 13B to 70B parameters designed to enhance the chat and reasoning capabilities of AI systems. OpenChat is particularly notable for its comprehensive evaluation and benchmarking, aiming to push the boundaries of what open-source language models can achieve in conversational AI.

LAMM

  • PROJECT DIRECT URL: https://openlamm.github.io/
  • DESCRIPTION: LAMM, pronounced as /læm/ and symbolizing a cute lamb, is at the forefront of multi-modal large language models (MLLMs) and applications. It represents a growing open-source community dedicated to enabling researchers and developers to train, evaluate, and build multi-modal AI agents. LAMM focuses on bridging the gap between ideas and execution, fostering seamless interaction between humans and AI machines through the development of datasets, frameworks, benchmarks, optimizations, and applications for MLLMs.

Platypus 2

  • PROJECT DIRECT URL: https://github.com/OthersideAI/Platypus
  • DESCRIPTION: Platypus is a fine-tuned and merged family of Large Language Models (LLMs) that has achieved top performance on the HuggingFace Open LLM Leaderboard. It utilizes a curated dataset, Open-Platypus, and a fine-tuning process that conserves the strong prior of pre-trained LLMs while enhancing domain-specific knowledge. The project focuses on optimizing LLMs using PEFT and LoRA with Open-Platypus, aiming for enhanced performance with minimal fine-tuning time and cost. Platypus models, including a 13B variant, have demonstrated strong performance across various quantitative LLM metrics, marking a significant contribution to the field of AI research.

Aviary

  • PROJECT DIRECT URL: https://github.com/ray-project/ray-llm
  • DESCRIPTION: Aviary is described as a new tool for the open-source LLM developer stack, released by Anyscale. It’s designed to help developers choose and deploy the right technologies for their LLM-based applications. For more information, please refer to Anyscale’s official website or their GitHub page​.

FinGPT

  • PROJECT DIRECT URL: https://github.com/AI4Finance-Foundation/FinGPT
  • DESCRIPTION: FinGPT is an open-source financial large language model aimed at advancing financial research, cooperation, and innovation. For detailed information and access to the project, please visit the official GitHub page of FinGPT or search for it on Arxiv for the associated research paper​.