bet0x

AI_Research

My Gen AI research

dataset

darija <-> english dataset

every-chatgpt-gui

Every front-end GUI client for ChatGPT

fastapi_tritonserver

functionary

Chat language model that can interpret and execute functions/plugins

GOT-OCR2.0

Official code implementation of General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model

gpt-fast

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

Hermes-Function-Calling

HippoRAG

HippoRAG is a novel RAG framework inspired by human long-term memory that enables LLMs to continuously integrate knowledge across external documents.

infinity

Infinity is a high-throughput, low-latency REST API for serving vector embeddings, supporting a wide range of text-embedding models and frameworks.

k2-data-prep

k2-train

Liger-Kernel

Efficient Triton Kernels for LLM Training

llm-text-completion-finetune

Guide on text completion large language model fine-tuning, including example scripts and training data acquiring.

lloco

The official repo for "LLoCo: Learning Long Contexts Offline"

marker

Convert PDF to markdown quickly with high accuracy

milvus_cli

Milvus Command Line

omniparse

Ingest, parse, and optimize any data format ➡️ from documents to multimedia ➡️ for enhanced compatibility with GenAI frameworks

openagi

Paving the way for open agents and AGI for all.

openai_trtllm

OpenAI compatible API for TensorRT LLM triton backend

ray_vllm_inference

A simple service that integrates vLLM with Ray Serve for fast and scalable LLM serving.

self-rag

This includes the original implementation of SELF-RAG: Learning to Retrieve, Generate and Critique through self-reflection by Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi.

tensorrt-test

TensorRT LLM Benchmark Configuration

tidb.ai

A [WIP] out-of-the-box RAG (Retrieval-Augmented Generation) app based on the [WIP] vector storage in TiDB Serverless.

TinyAgent

tiptap

The headless rich text editor framework for web artisans.

TPI-LLM

TPI-LLM: A High-Performance Tensor Parallelism Inference System for Edge LLM Services.

unsloth

5X faster 60% less memory QLoRA finetuning

Verba

Retrieval Augmented Generation (RAG) chatbot powered by Weaviate

workbench-example-mistral-finetune

An NVIDIA AI Workbench example project for fine-tuning a Mistral 7B model