bilelomrani1

hyperfine

A command-line benchmarking tool

aiosqlite

asyncio bridge to the standard sqlite3 module

aiolimiter

An efficient implementation of a rate limiter for asyncio.

distilabel

⚗️ distilabel is a framework for synthetic data and AI feedback for AI engineers that require high-quality outputs, full data ownership, and overall efficiency.

pathway

Python ETL framework for stream processing, real-time analytics, LLM pipelines, and RAG.

llm-course

Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

TorchCP

A Python toolbox for conformal prediction research on deep learning models, using PyTorch.

gradnorm-pytorch

A practical implementation of GradNorm, Gradient Normalization for Adaptive Loss Balancing, in Pytorch

ScaleLLM

A high-performance inference system for large language models, designed for production environments.

tokencost

Easy token price estimates for 400+ LLMs. TokenOps.

ATLAS

A principled instruction benchmark on formulating effective queries and prompts for large language models (LLMs). Our paper: https://arxiv.org/abs/2312.16171

PowerInfer

High-speed Large Language Model Serving on PCs with Consumer-grade GPUs

pychatml

Chat Markup Language conversation library

confection

:candy: Confection: the sweetest config system for Python

ml-sigma-reparam

ComfyUI

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

promptbase

All things prompt engineering

client-python

Python client library for Mistral AI platform

mixtral-inference

inference code for mixtral-8x7b-32kseqlen

megablocks

SDV

Synthetic data generation for tabular data

mlx

MLX: An array framework for Apple silicon

uncertain_ground_truth

Dermatology ddx dataset, Jax implementations of Monte Carlo conformal prediction, plausibility regions and statistical annotation aggregation from our recent work on uncertain ground truth (TMLR'23 and ArXiv pre-print).

adapters

A Unified Library for Parameter-Efficient and Modular Transfer Learning

scikit-network

Graph Algorithms

data-juicer

A one-stop data processing system to make data higher-quality, juicier, and more digestible for (multimodal) LLMs! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷为大模型提供更高质量、更丰富、更易”消化“的数据！

galactic

data cleaning and curation for unstructured text

unsloth

Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory

gpt-fast

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

marker

Convert PDF to markdown quickly with high accuracy