Codebase for Instruction Following without Instruction Tuning

A programming framework for agentic AI. Discord: https://aka.ms/autogen-dc. Roadmap: https://aka.ms/autogen-roadmap

LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning

Code for Husky, an open-source language agent that solves complex, multi-step reasoning tasks. Husky v1 addresses numerical, tabular and knowledge-based reasoning tasks.

Code for the paper "PIGEON: Predicting Image Geolocations".

a state-of-the-art-level open visual language model | 多模态预训练模型

Web UI for Tork Workflow Engine

🐚 OpenDevin: Code Less, Make More

A distributed workflow engine

DHS 2023 LLM Workshop by Sourab Mangrulkar

This is an official PyTorch implementation of our NeurIPS 2023 paper "GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization"

Evaluation and Tracking for LLM Experiments

Positional Skip-wise Training for Efficient Context Window Extension of LLMs to Extremely Length

Whisper realtime streaming for long speech-to-text transcription and translation

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

YaRN: Efficient Context Window Extension of Large Language Models

Rectified Rotary Position Embeddings

Generate instructions with Claude and ChatGPT

Teaching LLMs memory management for unbounded context 📚🦙

🐟 Code and models for the NeurIPS 2023 paper "Generating Images with Multimodal Language Models".

Implementation of paper "LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models"

Open source libraries and APIs to build custom preprocessing pipelines for labeling, training, or production machine learning pipelines.

Fine-tune mistral-7B on 3090s, a100s, h100s

Military Symbols in JavaScript

Efficient long-context fine-tuning, supervised fine-tuning, LongQA dataset.

Large Context Transformers