No need to say much about Large Language Models (LLM). They are the power behind tools like ChatGPT and Google's Bard. However, it's difficult to keep track of all the releases of models.
Several websites have compiled lists and databases of available LLMs. Each site has different levels of maintenance and varying data collected for each LLM. To compare LLMs, you may consider factors like Model Size, Pre-training Data, Fine-tuning Data, Supported Tasks, Model Output Quality, Resources and Documentation, and License (Open-source or commercial). We reviewed several online websites and pages that maintain LLM lists and gathered the best ones.
Considering editing prompts? Check out LLMStudio by TensorOps
LLM Garden by Superwise and partners is the best option to find the most updated list of LLMs. Maintained by major contributors the website allows submitting new reports to easily update the always refreshing list.
AlpacaEval is an LLM-based automatic evaluation tool that provides a fast, cost-effective, and reliable way to assess language models. It utilizes the AlpacaFarm evaluation set, which tests models' ability to follow general user instructions. By comparing the responses generated by models with reference Davinci003 responses using auto-annotators, AlpacaEval determines win rates and presents them on the leaderboard. The tool demonstrates a high agreement rate with human annotations, making it a valuable resource for benchmarking models. AlpacaEval actively encourages community contributions, allowing new models to be added to the leaderboard by following specific steps. Additionally, the platform welcomes the creation of new evaluators and evaluation sets, particularly for more complex instructions and safety testing.
As opposed to the other lists, AlpacaEval doesn't elaborate metadata details on the models, however those can be found in the links that lead to the models' home pages.
The Practical Guides for Large Language Models" is a curated and continuously updated list of practical guide resources focused on Large Language Models (LLMs). It is based on a survey paper titled "Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond" and the efforts of Xinya Du.
The survey paper covers various aspects of LLMs, particularly ChatGPT and its applications. The curated list includes resources that aim to assist practitioners in navigating the vast landscape of LLMs and their applications in natural language processing (NLP). These resources provide valuable insights into the usage restrictions of different LLMs and offer information on model and data licensing. The repository encourages users to utilize the listed resources and welcomes pull requests to enhance the quality and accuracy of the figure. If you find any resources helpful, citing the survey paper is recommended.
With 5.3K stars on GitHub Open LLMs stands out for it's very long list of LLMs and active community. Although the GitHub interface is not as good as Airtable, it is still quite easy to navigate though the table and find key insights about some very popular LLMs and some less familiar ones.
The GitHub page "Awesome-LLM" is a comprehensive resource that offers a curated list of various tools, libraries, and frameworks related to the field of Legal Language Modeling (LLM). It serves as a valuable reference for researchers, developers, and enthusiasts interested in exploring and working with language models specifically designed for legal applications. The page provides an extensive collection of links to repositories, papers, datasets, and projects that cover a wide range of topics, including legal text generation, natural language processing for law, legal knowledge extraction, and more. By compiling these resources, the Awesome-LLM page aims to facilitate collaboration, knowledge sharing, and advancement in the domain of legal language modeling.
The page is currently ranked high in the search for LLM lists however the level of detail and the comprehensiveness of the list is lacking compare to other pages.
Would you like to hear more about using LLMs in production? Don't hesitate to reach out to us with questions through this website!
Comments