An efficient, flexible and full-featured toolkit for fine-tuning LLM (InternLM2, Llama3, Phi3, Qwen, Mistral, ...)

经济学人(含音频)、纽约客、卫报、连线、大西洋月刊等英语杂志免费下载,支持epub、mobi、pdf格式, 每周更新

Implementation of Google's SELF-DISCOVER

FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation

AdalFlow: The “PyTorch” library to auto-optimize any LLM tasks.

Official code for "Block Transformer: Global-to-Local Language Modeling for Fast Inference"

Official JAX implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden States

Natty is a natural-language proof assistant with an embedded automatic prover for higher-order logic. It is in an early stage of development.

HippoRAG is a novel RAG framework inspired by human long-term memory that enables LLMs to continuously integrate knowledge across external documents. RAG + Knowledge Graphs + Personalized PageRank.

ChatPilot: Chat Agent Web UI，实现Chat对话前端，支持Google搜索、文件网址对话（RAG）、代码解释器功能，复现了Kimi Chat(文件，拖进来；网址，发出来)。

A multimodal agent framework for solving complex tasks

Code for Adam-mini: Use Fewer Learning Rates To Gain More https://arxiv.org/abs/2406.16793

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

The Abstraction and Reasoning Corpus

TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients.

Simplified Masked Diffusion Language Model

Autoregressive Model Beats Diffusion: 🦙 Llama for Scalable Image Generation

Implementation for MatMul-free LM.

Dashboard starter using NextUI V2 and Nextjs.

Memory Mosaics are networks of associative memories working in concert to achieve a prediction task.

MoRA: High-Rank Updating for Parameter-Efficient Fine-Tuning

Hrrformer: A Neuro-symbolic Self-attention Model (ICML23)

This project extends the idea of the innovative architecture of Kolmogorov-Arnold Networks (KAN) to the Convolutional Layers, changing the classic linear transformation of the convolution to learna…

The simplest, fastest repository for training/finetuning medium-sized xLSTMs.