We test AI on everyday work, every week.
Meeting notes, customer replies, Excel formulas, summaries, and trap questions that catch made-up answers.
Latest result
As of 2026-10-17, Gemini 3.8 Flash made the fewest mistakes on our 10 everyday tasks, passing 35 of 35 checks. Qwen 3.8 27B passed 30 of 30 and gave no answer on 1 task. On the 5 trap questions, models slipped in 2 of 20 answers. The most common mistake was “invented” (2 times).