Setup and run a local LLM and Chatbot using consumer grade hardware.

Develop, evaluate and monitor LLM applications at scale

Bringing Language Models to the Most Resource Constrained Devices

a collection of tiny llms with usecases

Build a tiny LLM from scratch.

No description available

实现一个小型的LLM，包含LORA微调、web端部署、借鉴Deepseek的MOE架构、潜在注意力机制和简单的RAG功能，以学习大模型流程为主

Setup and run a local LLM and Chatbot using consumer grade hardware.

Train a tiny language model using pure C++ from scratch without any third-party libraries. CUDA is supported.

No description available

TinyLLM is a research project focused on developing and training compact, specialized language models using publicly available datasets from platforms like Hugging Face. The project aims to explore diverse architectures and build expertise toward creating a high-performance coding model rivaling the capabilities of claude code

Tiny devices, big ideas , from bits to brilliance

A repository for LLM Learners.

simple gui for openai api for brick/feature mobile phones compatible with j2me

从零开始的小型LLM

LLM inference engine

TinyLLM with Attention Residue Architecture - PyTorch implementation of Attention Residuals (AttnRes) — arXiv:2603.15031 by Kimi Team. Replaces fixed residual connections with learned depth-wise attention. Full AttnRes + Block AttnRes built from scratch with side-by-side training comparison.

This repo has all the code (well organized) i wrote while reading the book "Build a Large Language Model from Scratch"

TinyLLM project is a lightweight, privacy-focused AI inference system designed to run locally on standard laptops without a dedicated GPU. It utilizes Microsoft's ONNX Runtime (`onnxruntime-genai`) and the `Phi-3.5-mini-instruct` model optimized for CPU performance.

Most tinyestllm ever created

An LLM model with 10 million parameters was designed for research purposes.

A minimal LLM implemented in pure Ruby

easy to understand

Edge LLM for In-Vehicle Deployment - Optimized small language model for automotive conversational AI

No description available

Code Repository for Uds-Pretraining LLM Software project

No description available

A simple inference engine written in Rust

Character-level Transformer LLM built from scratch in PyTorch.

No description available