Anthropic 的 Claude 3.7 表现最好，其次是 Claude 3.5。遗憾的是，谷歌的 Gemini 1.5 Pro 和 OpenAI 的 GPT-4o 表现不佳。 有趣的是，尽管 OpenAI 的 GPT-4o 等推理模型 ...

2025-03-03 13:10发布于北京新智元官方账号 【新智元导读】Karpathy发出灵魂拷问，评估AI究竟该看哪些指标？答案或许就藏在经典游戏里！最近，加州 ...

Karpathy发出灵魂拷问，评估AI究竟该看哪些指标？答案或许就藏在经典游戏里！最近，加州大学圣迭戈分校Hao AI Lab用超级马里奥等评测AI智能体，Claude ...

有趣的是，测试结果显示，Anthropic的Claude 3.5 Sonnet在「赚钱」能力上竟然超越了OpenAI自家的GPT-4o和o1模型。 昨天马斯克刚刚发布了号称「地表最聪明 ...

DuckDuckGo, a search engine that protects user privacy and does not personalize searches, has released Duck.ai, an interface for AI chatbots, to the public. Anyone can chat with chat models such ...

When users tried out the new model through A/B testing, satisfaction was roughly on par with Claude 3.5 Sonnet and significantly higher than similar models like GPT-4o mini and Claude 3.5 Haiku ...