Designing Silicon Brains using LLM: Leveraging ChatGPT for Automated Description of a Spiking Neuron Array

Michael Tomlinson, Joe Li, Andreas Andreou
1/25/2024
cs.ARcs.AIcs.ET

No official code found. Best implementation: Andreou-JHULabOrg/tinytapeout_05_chatgpt_snn (4 stars, Apache-2.0, last commit 2024-09-24).

Abstract

Large language models (LLMs) have made headlines for synthesizing correct-sounding responses to a variety of prompts, including code generation. In this paper, we present the prompts used to guide ChatGPT4 to produce a synthesizable and functional verilog description for the entirety of a programmable Spiking Neuron Array ASIC. This design flow showcases the current state of using ChatGPT4 for natural language driven hardware design. The AI-generated design was verified in simulation using handcrafted testbenches and has been submitted for fabrication in Skywater 130nm through Tiny Tapeout 5 using an open-source EDA flow.

View on arXivView PDF

Code Implementations(1)

Andreou-JHULabOrg/tinytapeout_05_chatgpt_snnCommunity100%
Apache-2.0

Open Source Tapeout with RTL generated with chatGPT 4

42Tcl, PythonNov 1, 20232 years agoApache-2.0

Similar Papers

VeriGen: A Large Language Model for Verilog Code Generation

86% match2023

Data is all you need: Finetuning LLMs for Chip Design via an Automated design-data augmentation framework

85% match1 repo622024

Natural Language to Verilog: Design of a Recurrent Spiking Neural Network using Large Language Models and ChatGPT

85% match2024

Are LLMs Any Good for High-Level Synthesis?

83% match2024

VerilogEval: Evaluating Large Language Models for Verilog Code Generation

83% match1 repo4542023

Automatically Improving LLM-based Verilog Generation using EDA Tool Feedback

83% match2024

Cite this paper

@article{tomlinson2024designing,
  title  = {Designing Silicon Brains using LLM: Leveraging ChatGPT for Automated Description of a Spiking Neuron Array},
  author = {Michael Tomlinson and Joe Li and Andreas Andreou},
  year   = {2024},
  eprint = {2402.10920},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2402.10920v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2402.10920.svg)](https://codeofpaper.com/paper/2402.10920)

Discussion