MCP简体中文Upgrade
Market/Categories/Developer & AI Platform/Captum

Independent product intelligence · Worldwide · 2026-05-01 → 2026-07-31

Captumcaptum.ai

Captum is an open-source, extensible library built on PyTorch for model interpretability. It supports various modalities and offers tools for understanding and analyzing PyTorch models.

Developer & AI Platformcaptum.aiTracked since 2026-08-211 discovery source
3KBacklinks
577Referring domains
26Domain authority
+0.9%Referring domains · 90 days

Authority & distribution

How strong is Captum’s web footprint?

Backlinks measure accumulated distribution; referring domains show how broadly that authority is spread. Both are observable on the open web, so both are published for every product here.

Backlinks

3KCurrent snapshot

Referring domains

577Current snapshot

Authority

26Domain authority · category median 22

90-day move

+0.9%Referring domains, last 90 days
At 26, Captum sits above the 22 median for its category.

Trajectory · Pro

The counts above are the latest of 370 daily observations. The curve through them is on the full profile.

See the trajectoryOpen the full profile

Traffic & engagement

Where Captum ranks

Measured · Pro

Measured for this product: global rank, country rank and category rank.

Unlock this profileOpen the full profile

AI referral visibility

Which AI assistants are already sending Captum traffic?

Measured · Pro

Measured for this product: AI referral traffic, which assistants refer it and which pages those referrals land on.

Unlock this profileOpen the full profile

Product profile

What Captum is selling.

Traffic makes more sense when you can connect it to the actual product promise.

“Captum”

Captum is an open-source, extensible library built on PyTorch for model interpretability. It supports various modalities and offers tools for understanding and analyzing PyTorch models.

Developer & AI PlatformSeen 2026-08-21 → 2026-08-21Coverage tier C

What we have measured

4 of 8 signal groups are available for this product: identity, authority and backlinks, AI referrals, search demand.

Not measured for this domain: traffic and engagement, audience demographics, country distribution, acquisition mix. Those sections are left out of the page rather than filled with estimates.

Canonical website: https://captum.ai

Observed across 1 discovery source.

Research questions

What the evidence answers about Captum.

Every answer is tied to a source we collected, and the questions with no evidence behind them are left unanswered rather than guessed.

What key features does Captum offer?

Captum's key features include multi-modal support for interpretability across vision, text, and more, compatibility with most PyTorch models requiring minimal modification, and an extensible, open-source library design for research.

How can you install Captum?

Captum can be installed via conda (conda install captum -c pytorch) or via pip (pip install captum).

What is Captum used for, and what problem does it solve?

Captum is an open-source library for model interpretability, built on PyTorch. It is used to understand and analyze the behavior of PyTorch models across various modalities like vision and text, solving the problem of making complex neural networks more interpretable.

Who is Captum designed for?

Captum is designed for developers and researchers working with PyTorch models who need to debug, explain, or analyze their models' decisions.

What is Captum?

Captum is an open-source, extensible library built on PyTorch for model interpretability.

5 of 6 research questions are answered for this product. The rest need source evidence we have not collected yet, so they are left unanswered rather than guessed.

Behind the lock

Unlock full Captum intelligence.

  • Backlink trajectoryThe figures above are the latest observation. The daily curve behind them holds 370.
  • AI referral detailHow much traffic assistants send, which 3 of them do it, and the 3 pages they land on.
  • Search structure and ranking depthOrganic against paid, brand against non-brand, and how many keywords rank in the top three against the long tail — which is what decides whether the demand is portable or tied to the name.
  • The rest of the marketPage through every leaderboard 100 rows at a time, put any five products side by side, and query the whole dataset over API / MCP.

Worldwide estimates · Data period 2026-05-01 → 2026-07-31. Figures are measured estimates intended for market research. Signals we have not measured for this domain are omitted, never estimated; a measured zero is still reported as zero. How this was measured