Ajmal K's Projects
:zap: Dynamically generated stats for your github readmes
Trying out MySQL with go.
Basic Http server in go
Learning how to build JSON APIs in go.
API for the GPT-J language model š¦. Including a FastAPI backend and a streamlit frontend
An implementation of model parallel GPT-2 and GPT-3-style models using the mesh-tensorflow library.
Create UIs for your machine learning model in Python in 3 minutes
A model detecting guns in an image
A hash table in C
a "hello, world" video codec for those new to video codecs
Http proxy server in Rust with hyper
JARVIS, a system to connect LLMs with ML community
Learning Triton by OpenAI
āØ My portfolio built with Next.js, Tailwind CSS, PlanetScale, and Vercel.
End-to-end stack for WebRTC. SFU media server and SDKs.
Easy-to-use LLM fine-tuning framework (LLaMA, BLOOM, Mistral, Baichuan, Qwen, ChatGLM)
Port of Facebook's LLaMA model in C/C++
Inference Llama 2 in one file of pure C
port of https://github.com/DavidAnson/markdownlint
Model parallel transformers in JAX and Haiku
A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API
A cloud-native vector database, storage for next generation AI applications
Minimalistic UI for Ollama LMs - This powerful react interface for LLMs drastically improves the chatbot experience and works offline.
Reference implementation of Mistral AI 7B v0.1 model.
A high-performance serving framework for ML models, offers dynamic batching and multi-stage pipeline to fully exploit your compute machine
The simplest, fastest repository for training/finetuning medium-sized GPTs.