College students prefer Gemini over ChatGPT and Claude for AI-assisted essay writing, according to blind test data from Study Arena. Analyzing 6,851 votes, Gemini secured a 39.6% choice rate, followed by Claude at 31.8% and ChatGPT/OpenAI at 29.2%. In blind comparisons, students selected responses without knowing the model behind them, ensuring choices were based purely on quality.

Gemini's edge came from balancing completeness, structure, and clarity—key traits in college-level writing. Longer responses also won more often, with the selected answers averaging 37% longer than alternatives. However, increasing reasoning settings did not improve performance; low and medium effort settings scored higher than high-effort ones.

Each model excelled in specific areas: Gemini led in writing feedback (41.7%), Claude in assignment planning (43.2%), and ChatGPT in research support (39.3%). The study recommends using models as editorial tools rather than replacements, with Gemini as the default starting point for refining drafts.