LLM training in simple, raw C/CUDAllm.c LLM training in simple, pure C/CUDA. There is no need for 245MB of PyTorch or 107MB of cPython. For example, training GPT-2 (CPU, fp32) is ~1,000 lines of clean code in a single file. It compiles and runs instantly, and exactly matches the PyTorch reference implementation. I chose GPT-2 as the first working example because it is the grand-daddy of LLMs, the first time the modern stack was put together. Currently, I am working on: direct CUDA implementati…
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation 🔥 InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation Haofan Wang* · Matteo Spinelli · Qixun Wang · Xu Bai · Zekui Qin · Anthony Chen InstantX Team *corresponding authors InstantStyle is a general framework that employs two straightforward yet potent techniques for achieving an effective disentanglement of style and content from reference images. Principle Separating Content from Image.…
Free, simple, and intuitive online database design tool and SQL generator. Free, simple, and intuitive database design tool and SQL generator. drawDB · Discord drawDB DrawDB is a robust and user-friendly database entity relationship (DBER) editor right in your browser. Build diagrams with a few clicks, export sql scripts, customize your editor, and more without creating an account. See the full set of features here. Getting Started Local Development git clone https://gith…
Generative AI suite powered by state-of-the-art models and providing advanced AI/AGI functions. It features AI personas, AGI functions, multi-model chats, text-to-image, voice, response streaming, code highlighting and execution, PDF import, presets for developers, much more. Deploy on-prem or in the cloud.BIG-AGI 🧠✨ Welcome to big-AGI, the AI suite for professionals that need function, form, simplicity, and speed. Powered by the latest models from 12 vendors and open-source servers, big-AGI of…
C++ Insights - See your source code with the eyes of a compiler C++ Insights - See your source code with the eyes of a compiler. Contents What Why Building Usage Get Involved Support What C++ Insights is a Clang-based tool that does a source-to-source transformation. The goal of C++ Insights is to make things visible that normally and intentionally happen behind the scenes. It's about the magic the compiler does for us to make things work. Take this piece of code for example: clas…
OpenAI API Free Reverse ProxyChatGPT API Free Reverse Proxy Welcome to the ChatGPT API Free Reverse Proxy project, a complimentary resource allowing seamless access to OpenAI's API. This project mirrors the official OpenAI API endpoints, enabling users to leverage OpenAI functionalities without direct cost. Dive into our documentation to discover how to set up your reverse proxy or connect with our hosted service for an even smoother experience. Quick Links Join our Discord Community for s…
FreeAskInternet is a completely free, private and locally running search aggregator & answer generate using LLM, without GPU needed. The user can ask a question and the system will make a multi engine search and combine the search result to the ChatGPT3.5 LLM and generate the answer based on search results.FreeAskInternet Running www.perplexity.ai like app complete FREE, LOCAL, PRIVATE and NO GPU NEED on any computer ImportantIf you are unable to use this project normally, it is most likely d…
PokéRogue is a browser based Pokémon fangame heavily inspired by the roguelite genre. Battle endlessly while gathering stacking items, exploring many different biomes, fighting trainers, bosses, and more! Contributing 🛠️ Development If you have the motivation and experience with Typescript/Javascript (or are willing to learn) please feel free to fork the repository and make pull requests with contributions. If you don't know what to work on but want to help, reference the below To-Do secti…
avail-light Light client for the Avail blockchain Introduction avail-light is a data availability light client with the following functionalities: Listening on the Avail network for finalized blocks Random sampling and proof verification of a predetermined number of cells ({row, col} pairs) on each new block. After successful block verification, confidence is calculated for a number of cells (N) in a matrix, with N depending on the percentage of certainty the light client wants to achie…