OpenLLM, an open platform for operating LLMs in production

🧰 Software Dev AI tools and trends

OpenLLM, an open platform for operating LLMs in production
OpenLLM is an open-source tool, allows developers to run, fine-tune, serve, and deploy any large language models (LLMs) with ease. It offers support for various models like StableLM, Falcon, Dolly, Flan-T5, ChatGLM, StarCoder, and more. Furthermore, it provides streamlined deployment options, flexible APIs, and options to build custom AI apps.

S.A.T.U.R.D.A.Y
S.A.T.U.R.D.A.Y is a toolbox for vocal computing. It provides tools to build elegant vocal interfaces to modern LLMs. The goal of this project is to foster a community of like-minded individuals who want to bring forth the technology we have been promised in sci-fi movies for decades.

GPT-Migrate: Easily migrate your codebase from one framework or language to another
GPT-Migrate aims to automatically migrate codebases from one framework or language to another. The project is still in development, but it has already been able to successfully migrate codebases from Python to JavaScript and from JavaScript to Python.

Microsoft's Semantic Kernel; Alternative to LangChain? 
Microsoft's Semantic Kernel is a lightweight SDK that allows developers to easily integrate LLMs into their applications. The extensible programming model combines natural language semantic functions and traditional code functions to add value to applications with AI. The SDK supports prompt templating, function chaining, vectorized memory, and intelligent planning capabilities out of the box. The project has been released as open-source.

I've yet to dive into this SDK but it does sound as thought there is a lot of overlap with Langchain, though at this point it looks like it only supports OpenAI LLMs.

Why should you combine ChatGPT with Knowledge Graphs?
Knowledge graphs provide a structured way of linking concepts together, allowing ChatGPT to make inferences and draw conclusions based on the information it has been provided, thus improving the reasoning capabilities of the model. Incorporating knowledge graph data into ChatGPT allows it to access a wealth of background information and context, allowing it to generate text with a greater depth of understanding and context.

Introducing the Vercel AI SDK
The Vercel AI SDK is an open-source library designed to help developers build conversational, streaming, and chat user interfaces in JavaScript and TypeScript. The SDK supports React/Next.js, Svelte/SvelteKit, with support for Nuxt/Vue coming soon.

GPT powered coding in your terminal

Aider is a command-line chat tool that allows you to write and edit code with OpenAI’s GPT models. You can ask GPT to help you start a new project, or modify code in your existing git repo. Aider makes it easy to git commit, diff & undo changes proposed by GPT without copy/pasting

Simple Chat — answer to your frustrations with LangChain
LangChain is a great library and community, but some might say of suffers greatly from attempting to keep pace with the rapid evolvement of LLMs and their APIs. The code and documentation can be difficult to follow at times. In this post Max Wolf vents on his frustration with LangChain, but finishes with the answer he has developed; SimpleAIChat.

Developer friendly, serverless vector database
LanceDB is a serverless vector database that makes data management for LLMs frictionless. It is lightweight and can be added into your app with no servers to manage. You can store, query and filter embeddings, metadata, text, video, audio and more

OpenSource Enterprise Question-Answering
Danswer is an open-source enterprise question-answering tool that allows you to ask natural language questions against internal documents and get back reliable answers backed by quotes and references from the source material so that you can always trust what you get back. It is powered by generative AI models with answers backed by quotes and source links. It also has intelligent document retrieval (semantic search/reranking pipeline) using the latest LLMs

LangChain, nearing 400 integrations

When working with LangChain, be sure to check what already exists prior to building something “new”. Will probably be over 400 integrations by the time you read this.

🔎 Focus on: Great Youtube channels for AI app builders

A great way to learn about the basics of AI Engineering is to watch it in action. I’ve found these YouTube channels immensely helpful for learning picking up core concepts. Most of these channels are demonstrating actual builds of apps and often links to the code is provided.

🏫 Read/Listen/Watch: Learn
Increase the breadth and depth of your AI knowledge

LabLab.AI
[read]
Dozens of tutorials on a wide array of Generative AI topics.

What is Sentence Similarity?
[Read or Watch]
Sentence embeddings are useful for search, translation, summarization and more. In this deep dive article from Cohere you are introduced to the primary topics and shown how they are utilized in LLMs

A Beginner's Guide to GANs
[read]
In this article learn about Generative Adversarial Networks (GANs) which are algorithmic architectures that use two neural networks, pitting one against the other (thus the “adversarial”) in order to generate new, synthetic instances of data that can pass for real data. They are used widely in image generation, video generation and voice generation.