How does the Hacadly AI Agent perform compared to Claude?

In WOGO's benchmark of 23 drawings, the Hacadly AI Agent had a 100% reconstruction success rate. A Claude-based system with tool integration succeeded on 11 of 23 (47.8%), and without tool integration on 9 of 23 (39.1%). Both used Claude Opus 5.5.

Can the Hacadly AI Agent reconstruct editable feature history from CAD data?

Yes. It rebuilds a feature history with sketches, extrusions, rotations, and holes from STEP data lacking feature history. In an edit evaluation, 33 of 34 cases passed (97.1%), meaning parameter changes could modify specified areas while preserving other shapes.

How is the Hacadly AI Agent deployed?

It is provided as software installed on each user's device. For enterprises, it can connect to a generative AI environment in the customer's own cloud, so data does not pass through WOGO's servers.

Large Language ModelsAI Business & IndustryITmedia AI＋13h agoNov 6, 2026, 19:01 JST

WOGO launches Hacadly AIエージェント for 2D-to-3D CAD automation

Send on LINE

3 Key Points

What happened
WOGO began offering Hacadly AI Agent on November 5, 2026. It reads DXF/PDF drawings and outputs STEP 3D models, passing shape checks on all 23 test drawings versus 11 of 23 (47.8%) for a Claude-based system with tool integration.
Why it matters
The agent lets designers turn existing 2D drawings into editable 3D models without manual remodeling, potentially saving time on derived designs and dimension changes.

WHO IT HITSMechanical design and CAD engineering teams who need to reuse legacy 2D drawings or edit CAD models without feature history may be able to automate conversion and reconstruction tasks, reducing manual remodeling work.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

WOGO's entry into AI-assisted mechanical design builds on its existing Hacadly series, which already includes 3D automated drawing-check and 2D automated drafting tools. The new agent is positioned to connect with those products, aiming to support a continuous workflow from reusing existing design assets to generating, modifying, and verifying models, and ultimately creating drawings.

The benchmark results suggest that combining a general-purpose LLM with WOGO's proprietary 2D drawing and 3D shape processing technology can outperform a Claude-based system alone, particularly when tool integration is absent. This indicates that domain-specific shape processing remains a differentiator even as foundation models improve.

Looking ahead, WOGO plans to expand coverage to more part shapes, assemblies, and CAD environments, and to deepen integration with its other Hacadly products. The company also offers deployment support, including task selection, validation with real data, CAD integration, and building internal generative AI environments, which may lower adoption barriers for enterprises.

FAQ

How does the Hacadly AI Agent perform compared to Claude?: In WOGO's benchmark of 23 drawings, the Hacadly AI Agent had a 100% reconstruction success rate. A Claude-based system with tool integration succeeded on 11 of 23 (47.8%), and without tool integration on 9 of 23 (39.1%). Both used Claude Opus 5.5.
Can the Hacadly AI Agent reconstruct editable feature history from CAD data?: Yes. It rebuilds a feature history with sketches, extrusions, rotations, and holes from STEP data lacking feature history. In an edit evaluation, 33 of 34 cases passed (97.1%), meaning parameter changes could modify specified areas while preserving other shapes.
How is the Hacadly AI Agent deployed?: It is provided as software installed on each user's device. For enterprises, it can connect to a generative AI environment in the customer's own cloud, so data does not pass through WOGO's servers.

ITmedia AI＋Read Original Article

Send on LINE

Also reported by AI Watch (Impress)

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.