反超DeepSeek！新版GPT-4o登顶竞技场，奥特曼：还会更好

The newly updated GPT-4o has surpassed DeepSeek-R1 to achieve a tie for first place in the large model arena. It excels in several categories, including creative writing, programming, and instruction adherence, while ranking sixth in mathematics.

Key Points

• GPT-4o now ranks first in creative writing, programming, and instruction following.
• DeepSeek-R1 has been surpassed in the large model benchmark rankings.
• GPT-4o ranks sixth in mathematics performance.
• The update indicates ongoing improvements in AI model capabilities.
• This development may influence future AI applications and competition.