Zero of fourteen models named Huntress Managed EDR first on the direct prompt; zero named Microsoft Security Copilot. Huntress Managed EDR was named by nine of the fourteen models and Microsoft Security Copilot by twelve and Huntress Managed EDR carries 11 labels and Microsoft Security Copilot 30, so the shares are not directly comparable.
Named in seven categories this edition.
Named in two categories 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 AI SOC agents page.
| Model | Direct | Paraphrase | Comparative | Budget-constrained | Scale-constrained | Negative |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | ||||||
| GPT-5.4 mini | MS | MS | MS | |||
| Gemini 3.5 Flash | MS | MS | MS | |||
| Perplexity Sonar | MS | MS | MS | MS | ||
| Grok 4.20 | MS | MS | ||||
| Mistral Small | MS | MS | MS | |||
| DeepSeek V4 Flash | MS | MS | MS | |||
| Llama 4 Maverick | ||||||
| Qwen 3.7 Flash | MS | MS | ||||
| Kimi K2 | MS | MS | ||||
| GLM 4.7 FlashX | MS | MS | ||||
| MiniMax M2.5 | ||||||
| GPT-6 Luna | MS | |||||
| Muse Glimmer 30B | MS | MS |
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. One of one in this category shown.
“the most cost-predictable commercial agentic option in the current evaluations is UnderDefense Agentic AI SOC or Huntress” Muse Glimmer 30B · budget prompt · first choice
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Five of five in this category shown.
“enterprise AI SOC solutions like Microsoft Security Copilot, ReliaQuest GreyMatter, or CrowdStrike Charlotte AI are often cost-prohibitive” Gemini 3.5 Flash · budget prompt · soft negative
“Microsoft Security Copilot — noted as having "several components still in preview" (feels incomplete in production)” DeepSeek V4 Flash · negative prompt · soft negative
“Limited effectiveness outside of Microsoft-native telemetry; currently more of a "chatbot" than a true agent.” Qwen 3.7 Flash · comparative prompt · soft negative
“I recommend Microsoft Security Copilot or Dropzone AI as your AI SOC analyst for alert triage” Mistral Small · paraphrase prompt · first choice
“Companies already standardized on Defender, Sentinel, Entra, and Microsoft 365 | Less compelling if your environment is vendor-diverse” Perplexity Sonar · paraphrase prompt · alternative
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.