dctanner

recomp

RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation.

sqlite-vec

Work-in-progress vector search SQLite extension that runs anywhere.

llmsherpa

Developer APIs to Accelerate LLM Projects

nlm-ingestor

This repo provides the server side code for llmsherpa API to connect. It includes parsers for various file formats.

LLMCompiler

[ICML 2024] LLMCompiler: An LLM Compiler for Parallel Function Calling

open-parse

Improved file parsing for LLM’s

sqlcoder

SoTA LLM for converting natural language questions to SQL queries

MetaGPT

🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

E2B

Secure cloud runtime for AI apps & AI agents. Fully open-source.

infra

Infrastructure powering E2B - Secure Runtime for AI Agents & Apps

supermemory

Build your own second brain with supermemory. It's a ChatGPT for your bookmarks. Import tweets or save websites and content using the chrome extension.

marker

Convert PDF to markdown quickly with high accuracy

zod-gpt

Get structured, fully typed, and validated JSON outputs from OpenAI and Anthropic models.

aici

AICI: Prompts as (Wasm) Programs

Flowise

Drag & drop UI to build your customized LLM flow

WeightWatcher

The WeightWatcher tool for predicting the accuracy of Deep Neural Networks

aphrodite-engine

PygmalionAI's large-scale inference engine

moondream

tiny vision language model

llm

Access large language models from the command-line

nanotron

Minimalistic large language model 3D-parallelism training

dspy

DSPy: The framework for programming—not prompting—foundation models

NeuScraper

This is the code repo for our paper "Cleaner Pretraining Corpus Curation with Neural Web Scraping".

min-max-gpt

Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training

google-indexing-script

Script to get your site indexed on Google in less than 48 hours

cosmopedia

vram-calculator

Transformer GPU VRAM estimator

datatrove

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

LLaMA-Factory

Unify Efficient Fine-Tuning of 100+ LLMs

101-school

AI generated courses

tinyllama

A side project that follows all the acceleration tricks in tinyllama, with the minimal modification to the huggingface transformers code.