4bit

FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.

Custom model training using modern architectures. 4-bit QLoRA fine-tuning pipeline for LLaMA 3 8B with production-grade optimization. Memory-efficient training on consumer GPUs. Published adapter on HuggingFace. From training pipeline to deployed model.

QuantLLM is a Python library designed for developers, researchers, and teams who want to fine-tune and deploy large language models (LLMs) efficiently using 4-bit and 8-bit quantization techniques.

Commented source code of Texas Instrument's original Speak and Spell™

A Raspberry Pi Pico (RP2040)-based 2114 SRAM Emulator for the Busch 2090 Microtronic Computer System

GPTQ optimized for Ampere (RTX 3090/3060) — Marlin JIT nvcc compatibility fixes included

Supporting code for "LLMs for your iPhone: Whole-Tensor 4 Bit Quantization"

Arduino library for 4-Bit LED Digital Tube Module with two 74HC595D-chip

These are VHDL codes for a signed 4bit multiplier using 4bit adders. Base on Baugh-Wooley Method.

This is a VHDL code for 4bit multiplier using 4bit full adder circuit structurally modelled.

Simulation of a 4-bit CPU with Logisim

Mixed-precision quantization scheme (16/8/4bit mixed quantization) for the Wan2.2-Animate-14B model. Compresses the original 35GB base model to 17GB, balancing inference performance and model size.

基于 Meta-Llama-3.1-8B-Instruct + 4-bit 量化 + QLoRA，训练与推理全程显存占用 < 9 GB，RTX 3060/4060 即可玩转。

Design Project for a 4-bit Micro-Processor for CS1050

This emulates the 4bit computer and be able to run on the browser.

📚 This repository demonstrates how to interface a 16x2 alphanumeric LCD with the 8051 Microcontrollers (AT89S52) using Assembly Language Programming. The project is designed to showcase a practical example of sending data from the 8051 microcontroller to an LCD and includes Proteus simulation files for testing, modify, debug and visualization.

a 4 bit TTL computer

Layout of 4bit Ripple Carry Adder formed using CMOS logic in gpdk180nm technology node done in Cadence Virtuoso with no DRC and LVS errors.

Fused 4bit AdamW in Cuda

Decodificador de pantalla de 7 segmentos Ánodo común en base a una entrada de 4bit escrito en VHDL.