**Benchmark for AI Assistants**

> A public scorecard for AI assistants you can text. Every assistant scored on the same 15 dimensions, backed by real public quotes. Scorecard 71 assistants, 15 dimensions, one scale. Overall is a running mean until an assistant is fully tested. Benchmark v 0.2 15 tasks 180 runs last test Sep 14, 2026 updated Sep 14, 2026 General 44 Travel 2 Email 3 Finance 3 Shopping 2 Health & food 1 Work & teams 24 Infra & hardware...

**来源信息**
- **来源**：Hacker News：AI 热帖
- **分类**：论文
- **发布时间**：2026-09-15 13:50（北京时间）
- **原文**：[打开原文](https://assistantbenchmark.com/)