Two of fourteen models named Constructor first on the direct prompt; two named Klevu. Constructor was named by fourteen of the fourteen models and Klevu by eleven and Constructor carries 42 labels and Klevu 28, 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 ecommerce search and merchandising page.
Across every category in the November 2026 Edition, Constructor and Klevu were named in the same answer forty-two times, of the 133 answers naming Constructor and the 70 naming Klevu. In those answers Klevu took the first choice one time and Constructor ten.
| Model | Direct | Paraphrase | Comparative | Budget-constrained | Scale-constrained | Negative |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | ||||||
| GPT-5.4 mini | ||||||
| Gemini 3.5 Flash | ||||||
| Perplexity Sonar | ||||||
| Grok 4.20 | ||||||
| Mistral Small | ||||||
| DeepSeek V4 Flash | ||||||
| Llama 4 Maverick | ||||||
| Qwen 3.7 Flash | ||||||
| Kimi K2 | ||||||
| GLM 4.7 FlashX | ||||||
| MiniMax M2.5 | ||||||
| GPT-6 Luna | ||||||
| Muse Glimmer 30B |
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 eight in this category shown.
“Overkill and expensive ($100s–$1,000s/month); not suitable for limited budgets.” Grok 4.20 · budget prompt · hard negative
“Bloomreach, Coveo, Constructor.io, Google Vertex AI Search for Commerce | SMBs & Mid-Market: ... Avoid them unless you have dedicated merchandising teams” Gemini 3.5 Flash · negative prompt · soft negative
“Constructor — strong product discovery, but ensure the team can handle the learning curve and workflow complexity.” GPT-5.4 mini · negative prompt · soft negative
“Best overall for most mid-market B2B: Constructor — Strongest in AI-driven search, personalization, and merchandising for complex B2B catalogs.” Grok 4.20 · direct prompt · first choice
“My first choice would be Constructor if your buyers see different catalogs, prices, or availability depending on their account.” GPT-6 Luna · paraphrase prompt · first choice
“Luigi's Box and Constructor are top choices due to their B2B-specific features and strong performance.” Mistral Small · direct 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.
“Enterprise tools like Klevu, Searchspring, or Constructor: Overkill and expensive” Grok 4.20 · budget prompt · hard negative
“Klevu — can be strong, but pricing/ops fit may vary, and you should test merchandising UX deeply.” GPT-5.4 mini · negative prompt · soft negative
“The combined cost can escalate quickly and becomes significantly higher than competitors” Qwen 3.7 Flash · negative prompt · soft negative
“Klevu is an ecommerce search and navigation platform built for mid-market brands... It provides fast setup, AI-enhanced semantic search, self-service merchandising” Claude Haiku 4.5 · direct prompt · first choice
“Start with Hawksearch or Klevu if you want a balance of ease-of-use, B2B-specific features, reasonable pricing, and quick implementation” Claude Haiku 4.5 · paraphrase prompt · first choice
“Klevu (now part of Athos Commerce) ⭐ *Best Overall for Mid-Market*” Kimi K2 · direct 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.