ericxsun

WizardLM

LLMs build upon Evol Insturct: WizardLM, WizardCoder, WizardMath

streaming-llm

[ICLR 2024] Efficient Streaming Language Models with Attention Sinks

lit-gpt

Hackable implementation of state-of-the-art open-source LLMs based on nanoGPT. Supports flash attention, 4-bit and 8-bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

Memex

Browser extension to curate, annotate, and discuss the most valuable content and ideas on the web. As individuals, teams and communities.

llm-foundry

LLM training code for Databricks foundation models

h2o-llmstudio

H2O LLM Studio - a framework and no-code GUI for fine-tuning LLMs. Documentation: https://h2oai.github.io/h2o-llmstudio/

localpilot

leptonai

A Pythonic framework to simplify AI service building

vdp

💧 Instill VDP (Versatile Data Pipeline) is an open-source tool to seamlessly integrate AI to process unstructured data in the modern data stack

state-of-open-source-ai

:closed_book: Clarity in the current fast-paced mess of Open Source innovation

prompts-for-edu

BitNet

Implementation of "BitNet: Scaling 1-bit Transformers for Large Language Models" in pytorch

Skywork

Skywork series models are pre-trained on 3.2TB of high-quality multilingual (mainly Chinese and English) and code data. We have open-sourced the model, training data, evaluation data, evaluation methods, etc. 天工系列模型在3.2TB高质量多语言和代码数据上进行预训练。我们开源了模型参数，训练数据，评估数据，评估方法。

modelzoo

local-persist

Create named local volumes that persist in the location(s) you want

rl_a3c_pytorch

A3C LSTM Atari with Pytorch plus A3G design

textbook_quality

Generate textbook-quality synthetic LLM pretraining data

carton

Run any ML model from any programming language.

adept-inference

Inference code for Persimmon-8B

libgen_to_txt

Convert all of libgen to high quality markdown

pynvml

Provide Python access to the NVML library for GPU diagnostics

AttentionIsOFFByOne

Implementation of "Attention Is Off By One" by Evan Miller

multipack_sampler

Multipack distributed sampler for fast padding-free training of LLMs

smart_agent

AutoNetGen

让 AI 设计 AI，让大模型帮助小模型进化，用魔法创造魔法！ Empower Artificial Intelligence to sculpt its own kind, where colossal models gracefully usher the petite ones into evolution, weaving magic to conjure further enchantment!

Nanbeige

ADaPT

Official code for the paper "ADaPT: As-Needed Decomposition and Planning with Language Models"

Omnigrok

Omnigrok: Grokking Beyond Algorithmic Data

colorbindiff

A visual and colorized diff for binary files.

tqdm-loggable

Logging friendly progress messages for TQDM progress bars