bevren

langroid

Harness LLMs with Multi-Agent Programming

LLM101n

LLM101n: Let's build a Storyteller

hallo

Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation

Loyal-Elephie

Your Trusty Memory-enabled AI Companion - Simple RAG chatbot optimized for local LLMs | 12 Languages Supported | OpenAI API Compatible

mistral-inference

Official inference library for Mistral models

LLM-RLHF-Tuning-with-PPO-and-DPO

Comprehensive toolkit for Reinforcement Learning from Human Feedback (RLHF) training, featuring instruction fine-tuning, reward model training, and support for PPO and DPO algorithms with various configurations for the Alpaca, LLaMA, and LLaMA2 models.

abliterator

Simple Python library/structure to ablate features in LLMs which are supported by TransformerLens

llama.np

Inference Llama/Llama2 Modes in NumPy

llama3.cp

Adapted version of llama3.np (NumPy) to a CuPy implementation for the Llama 3 model.

GlaDOS

This is the Personality Core for GLaDOS, the first steps towards a real-life implementation of the AI from the Portal series by Valve.

llama.cpp

LLM inference in C/C++

stable-diffusion.cpp

Stable Diffusion in pure C/C++

MeloTTS

High-quality multi-lingual text-to-speech library by MyShell.ai. Support English, Spanish, French, Chinese, Japanese and Korean.

onnxruntime

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

seemore

From scratch implementation of a vision language model in pure PyTorch

Scrapegraph-ai

Python scraper based on AI

outlines

Structured Text Generation

Cherry_LLM

[NAACL'24] Self-data filtering of LLM instruction-tuning data using a novel perplexity-based difficulty score, without using any other models

aici

AICI: Prompts as (Wasm) Programs

fsdp_qlora

Training LLMs with QLoRA + FSDP

GaLore

GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

bitnet

Modeling code for a BitNet b1.58 Llama-style model.

moondream

tiny vision language model

bgpt

Beyond Language Models: Byte Models are Digital World Simulators

Grounded-Segment-Anything

Grounded SAM: Marrying Grounding DINO with Segment Anything & Stable Diffusion & Recognize Anything - Automatically Detect , Segment and Generate Anything

Superfiltering

[ACL'24] Superfiltering: Weak-to-Strong Data Filtering for Fast Instruction-Tuning

graphcast

ringattention

Transformers with Arbitrarily Large Context

LWM

three-bmfont-text

renders BMFont files in ThreeJS with word-wrapping