tensorViz logo

tensorViz

Let your coding agent see inside your PyTorch model

Listing details

HiCyou directory record

Listed
Listed website
producthunt.com
Directory record created
Nov 21, 2026
Last record update
Nov 21, 2026

Listing information

  • Overview
  • 5 key features
  • 3 use cases

This page describes HiCyou's directory record. It is not a security, ownership, or product-quality certification.

Review the external site before sharing sensitive information or making a purchase.

tensorViz preview

What is tensorViz

tensorViz is a free, open-source (MIT) tool that gives coding agents visibility into PyTorch models. It runs a real forward pass and converts it into an interactive graph showing every module and tensor shape. AI coding tools like Claude Code, Codex, and Cursor can search this graph and read measured shapes, enabling them to fix model code based on evidence rather than guesses. Its Model Explorer also offers browsable graphs of 560+ models from recent papers.

Key Features

Runs one real forward pass to capture actual data flow through a PyTorch model
Generates an interactive graph with every module and tensor shape
Lets coding agents such as Claude Code, Codex, and Cursor search the graph and read measured shapes
Model Explorer with interactive graphs of 560+ models from recent papers
Free and open source under the MIT license

Use Cases

  • Debugging PyTorch model code with evidence from actual tensor shapes
  • Giving AI coding assistants ground truth about model architecture during code fixes
  • Browsing and discussing graphs of published models without installing anything

Why do startups need this tool?

Startups building with PyTorch can use tensorViz to make AI-assisted development more reliable: instead of agents guessing at model structure, they read measured shapes from a real forward pass. Being free and MIT-licensed, teams can adopt it without procurement friction, and the Model Explorer offers a quick way to study published architectures.

FAQs