Eight of fourteen models named SAP LeanIX first on the direct prompt; three named Ardoq. SAP LeanIX was named by twelve of the fourteen models and Ardoq by thirteen and SAP LeanIX carries 33 labels and Ardoq 47, so the shares are not directly comparable.
Named in two categories this edition.
Named in one category this edition.
Share is the count of first choices across the direct, paraphrase, budget and scale prompts over all fourteen models, for a mid-market B2B company; rank is within the category; every quote names the model and the prompt it came from. Both figures come from the enterprise architecture tools page.
Across every category in the October 2026 Edition, SAP LeanIX and Ardoq were named in the same answer ninety-five times, of the 119 answers naming SAP LeanIX and the 145 naming Ardoq. In those answers Ardoq took the first choice six times and SAP LeanIX twenty-nine.
| Model | DirectSL | ParaphraseSL | ComparativeSL | Budget-constrainedSL | Scale-constrainedSL | NegativeSL |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | ||||||
| GPT-5.4 mini | SL | SL | SL | |||
| Gemini 3.5 Flash | SL | SL | SL | |||
| Perplexity Sonar | SL | SL | ||||
| Grok 4.1 Fast | SL | SL | ||||
| Mistral Small | ||||||
| DeepSeek V4 Flash | SL | SL | SL | SL | SL | |
| Llama 4 Maverick | ||||||
| Qwen 3.7 Flash | SL | SL | ||||
| Kimi K2 | SL | SL | SL | |||
| GLM 4.7 FlashX | SL | |||||
| MiniMax M2.5 | SL | SL | ||||
| GPT-6 Luna | SL | SL | SL | |||
| Muse Glimmer 30B | SL | SL | SL |
Bold names in an answer are the products the judge labeled a first choice; a model naming several gives each of them that label. The full answer text for every row is in the record.
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of seven in this category shown.
“SAP LeanIX, if you need a highly bespoke metamodel... That's not a reason to avoid LeanIX if its standard model fits” GPT-6 Luna · negative prompt · soft negative
“Medium-High Risk: Customization & Data Model Rigidity ... LeanIX if you need deep customization” DeepSeek V4 Flash · negative prompt · soft negative
“Often targets larger enterprises; check if the licensing/price is competitive for <1k users.” Qwen 3.7 Flash · scale prompt · soft negative
“agile cloud-based solutions with strong integration and collaboration features are typically recommended, e.g. SAP LeanIX, Ardoq, or BlueDolphin” Muse Glimmer 30B · scale prompt · first choice
“Choose LeanIX if you want to quickly understand what applications you own, how much they cost, and how they map to business goals.” Qwen 3.7 Flash · comparative prompt · first choice
“LeanIX is highly accessible for mid-market companies due to its out-of-the-box usability and modular pricing structure” Gemini 3.5 Flash · paraphrase prompt · first choice
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“Avoid high-end tools like LeanIX, Bizzdesign, or Ardoq” Grok 4.1 Fast · budget prompt · hard negative
“High Risk: Pricing/Packaging Changes ... Ardoq if you have a very large application portfolio (>1,000 apps) and budget sensitivity.” DeepSeek V4 Flash · negative prompt · soft negative
“Ardoq is an example often flagged for this: ... exclusively a cloud-hosted SaaS platform — there is no on-premises deployment option” Muse Glimmer 30B · negative prompt · soft negative
“Dynamic visualization & linking complex relationships. Very user-friendly for non-architects. | Great for "living repository" approach.” Qwen 3.7 Flash · scale prompt · first choice
“Ardoq is the most cited "modern" option for teams that want browser-based real-time collaboration and automated dependency mapping.” Muse Glimmer 30B · direct prompt · first choice
“LeanIX and Ardoq are usually the strongest fits because they prioritize portfolio views, impact analysis, and change planning” Perplexity Sonar · comparative prompt · first choice
Comparisons are drawn for the top eight products in each category, each against each. The output is the models' output; nothing here is a recommendation by the index.