Zero of fourteen models named Guideline first on the direct prompt; four named Empower Retirement. Guideline was named by twelve of the fourteen models and Empower Retirement by thirteen and Guideline carries 25 labels and Empower Retirement 37, 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 401(k) providers page.
| 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. Four of five in this category shown.
“execution and service seem unreliable—avoid if you value responsive support.” Grok 4.20 · negative prompt · hard negative
“Guideline is frequently ranked first for cost/automation, with $89 base + $8/employee in current provider listings, and no setup fee.” Muse Glimmer 30B · budget prompt · first choice
“Choose a Tech-Forward Provider (Guideline, Human Interest, Betterment) if you have under 50 employees” Gemini 3.5 Flash · comparative prompt · first choice
“Guideline looks like the strongest low-budget option for many small businesses” GPT-5.4 mini · budget 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.
“Be most cautious with: Empower (service meltdowns)” Kimi K2 · negative prompt · hard negative
“the SEC found that Empower entities inadequately disclosed conflicts and financial incentives tied to recommendations to enroll participants in that service” GPT-6 Luna · negative prompt · soft negative
“remains a mid-tier provider where plan design (chosen by your employer) determines whether you get good or bad options” DeepSeek V4 Flash · negative prompt · soft negative
“Scale / investment choice + fiduciary support: Fidelity Workplace Investing, Empower Retirement” Muse Glimmer 30B · direct prompt · first choice
“Empower, Vanguard, Fidelity, and ADP are frequently cited as top choices” Mistral Small · direct prompt · first choice
“Why I'd lean Empower for a true mid-market B2B company” GPT-5.4 mini · 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.