量子位 QbitAIJun 8, 2026, 9:02 AM

ChatGPT vs Doubao on Gaokao Math

Original: 难上热搜的高考数学,我拿ChatGPT和豆包PK了一把!

QbitAI compares ChatGPT and Doubao using China’s Gaokao math questions.

The article appears to test ChatGPT and Doubao on Chinese Gaokao math problems. Since the original text is unavailable, the exact questions, prompts, scores, and winner cannot be verified. It should be treated as a media-style AI capability comparison rather than a rigorous, reproducible benchmark.

From the headline, this QbitAI article focuses on a topical test: using the gaokao (college entrance exam) math as a source of questions and pitting ChatGPT against Doubao in a head-to-head matchup. In the Chinese context, gaokao math typically represents a high-pressure, standardized exam format that requires multi-step reasoning, and is therefore often used by media to test large language models' abilities in problem-solving, reasoning, computation, and written explanation. However, because the original content was not provided, it is impossible to judge which questions the article actually used, whether they included multiple-choice or proof questions, whether multi-turn prompting was allowed, whether there was manual prompt correction, or to infer which model ultimately won.

Full summary

Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.

See Pro plans →

Want the original English / full article?

Read on 量子位 QbitAI →

Summaries are AI-generated; the original article is authoritative.