Jason3900

Open-Assistant

OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.

faiss

A library for efficient similarity search and clustering of dense vectors.

ChatGPT

Reverse engineered ChatGPT API

taichi

Productive, portable, and performant GPU programming in Python.

evals

Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.

RWKV-LM

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.

pytube

A lightweight, dependency-free Python library (and command-line utility) for downloading YouTube Videos.

xxHash

Extremely fast non-cryptographic hash algorithm

txtai

💡 All-in-one open-source embeddings database for semantic search, LLM orchestration and language model workflows

SagerNet

The universal proxy toolchain for Android

AutoGPTQ

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

P-tuning-v2

An optimized deep prompt tuning strategy comparable to fine-tuning across scales and tasks

DeepSpeed-MII

MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.

summarize-from-feedback

Code for "Learning to summarize from human feedback"

Watermark-Removal-Pytorch

🔥 CNN for Watermark Removal using Deep Image Prior with Pytorch 🔥.

calm

CUDA/Metal accelerated language model inference

Open-MAGVIT2

Open-MAGVIT2: Democratizing Autoregressive Visual Generation

few-shot-learning

Few-shot Learning of GPT-3

FlagPerf

FlagPerf is an open-source software platform for benchmarking AI chips.

latex-chinese-fonts

Simplified Chinese fonts for the LaTeX typesetting.

scripta-sinica

汉语古典文本资料库

FlagData

baukit

lm_perplexity

FlagBoot

fast-gector

CORT

Official code of EMNLP'22 paper "CORT: A New Baseline for Comparative Opinion Classification by Dual Prompts"

TGEA

ArabicText-NTG

EasyDistPLMs