Data as of Aug 25, 2026 · Based on 3,181,687 AI responses across 10,525 prompts · See how Parse measures this
Captum is an open-source, extensible library for model interpretability built on PyTorch. It supports interpretability across multiple modalities including vision and text, and can be used with most PyTorch models with minimal modification.
Words AI uses
AI reaches for open-source · interpretability library · dedicated when it describes Captum.
Sources
captum.ai shapes more of what AI says about Captum than any other source, at 10% of its citations.
medium.com · aiwiki.ai · mdpi.com · pmc.ncbi.nlm.nih.gov
The market map
Explainable AI & Model Monitoring Platforms →Where AI ranks Captum
Excerpts where Captum appeared in the AI's answer

Captum (Meta AI) — interpretability library for PyTorch models, including methods such as Integrated Gradients and saliency analysis.

Captum is an open-source, extensible library built on PyTorch by Meta for model interpretability and understanding, available on Captum.ai.
Excerpts where Captum appeared in the AI's answer

Captum (by PyTorch) - What it is: A production-ready model interpretability and feature attribution framework.
Excerpts where Captum appeared in the AI's answer

Captum: Designed for PyTorch models, Captum provides algorithms like Integrated Gradients to provide feature attributions.