n8n vs Tray.ai
Zero of fourteen models named n8n first on the direct prompt; zero named Tray.ai. n8n was named by fourteen of the fourteen models and Tray.ai by ten and n8n carries 48 labels and Tray.ai 19, so the shares are not directly comparable.
Berlin, Spain, founded 2019. Named in one category 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 workflow automation and ipaas page.
By framing
n8nFirst choices, of fourteen modelsTray.ai
Across every category in the October 2026 Edition, n8n and Tray.ai were named in the same answer twenty-two times, of the 131 answers naming n8n and the 47 naming Tray.ai. In those answers Tray.ai took the first choice three times and n8n one.
Every model, every framing
| 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 | | | | |
n8n Tray.ai first choice named as an alternative argued againstblank: not namedEach cell is one answer, n8n on the left and Tray.ai on the right.The direct prompt
Neither was the first choice, one was named
3 of 14 modelsThe answer put something else first and named one of the two as an alternative.Neither was named
11 of 14 modelsThe answer made no first choice from these two in this category.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.
By buyer segment
What the models said about n8n
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“Critical vulnerabilities: Multiple CVEs (CVE-2026-27577, CVE-2026-27493) with CVSS scores of 9.4-9.5” GLM 4.7 FlashX · negative prompt · hard negative
“Tens of thousands of systems may be exposed to a critical vulnerability in n8n (CVE-2026-21858)” Claude Haiku 4.5 · negative prompt · hard negative
“This popular open-source workflow automation platform has faced multiple critical vulnerabilities... If you use n8n, ensure you're running the latest versions” MiniMax M2.5 · negative prompt · soft negative
“For a company with a limited budget, n8n stands out as the best overall workflow automation and integration platform.” Grok 4.20 · budget prompt · first choice
“Best Overall Value (Free Option Available) ... n8n is unbeatable if you have technical resources to self-host” Kimi K2 · budget prompt · first choice
“n8n self-hosted or Activepieces gives you truly zero vendor cost and unlimited runs” Muse Glimmer 30B · budget prompt · first choice
What the models said about Tray.ai
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-scale complexity; pricing is prohibitive for mid-market.” Kimi K2 · direct prompt · hard negative
“not bad tools, but they are often enterprise-first and can require significant IT involvement” Perplexity Sonar · negative prompt · soft negative
“Avoid Workato or Tray.io if you have a tight budget or simple needs” Gemini 3.5 Flash · negative prompt · soft negative
“Workato and Tray.io tend to hit the sweet spot for this stage” DeepSeek V4 Flash · scale prompt · first choice
“Tools like Workato or Tray.io shine for your size” Grok 4.20 · scale prompt · first choice
“The Best Overall Choice: Tray.io” Qwen 3.7 Flash · paraphrase 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.