Is GPT 5 Mini good at Executive Assistant?
GPT 5 Mini ranks #48 of 50 for Executive Assistant — usable. The top pick for this task is claude-opus-4.5-high.
GPT 5 Mini on each Executive Assistant sub-task
| Tactful Rewrite Test | 86.7/100 | #9 |
| Priority Triage Test | 69.7/100 | #44 |
| Useful in Five Minutes | 62.3/100 | #50 |
| Message Risk Review | 62.0/100 | #50 |
Real examples, graded
WinClient scope creep response 94/100
“The response perfectly executes the prompt's instructions. It maintains the boundary firmly without being aggressive, offers concrete and realistic next steps, avoids apologizing entirely, and is highly concise. The tone is professional, calm, and production-ready.”
WeakOverpromising sales reply 62/100
“The model successfully identifies risks and provides safe rewrites, but fails on concision and commercial judgment by including a massive, generic software integration playbook (mentioning idempotency, canary rollouts, etc.) instead of a focused sales strategy.”
WeakInternal conflict meeting 67/100
“The model provides highly actionable, realistic, and well-judged advice that perfectly addresses the tension and facts. However, it includes several unrequested sections that add unnecessary length for a 5-minute preparation scenario, impacting concision and prioritization.”
Frequently asked
Is GPT 5 Mini good at Executive Assistant?
GPT 5 Mini ranks #48 of 50 models we tested for Executive Assistant, scoring usable.
What is GPT 5 Mini's strongest Executive Assistant skill?
Its best sub-task here is Tactful Rewrite Test.
This page is Spring Prompt, running
We just did this for every model. Do it for your prompt.
The rankings above come from running real tasks through real models and scoring every output. Spring Prompt is that same engine — pointed at your prompt, your test cases, and your definition of good.
- Generate test cases from your prompt — no eval set required to start.
- Compare models side by side with quality, cost and latency in one matrix.
- Optimise the winner until the scores say it's ready to ship.
Prompt × model results
12 test cases · 3 evals