还有一个重要的指标——准确率。伯克利函数调用排行榜 (BFCL) 是评估函数调用能力的标准基准。 Gemma 3 1B 的得分约为 31%，Llama 3.2 1B 约为 26%，两者未经微调的性能都很弱。由于 Gemma 3n 是通用型程序，因此未对其进行测试。Hammer 2.1 0.5B 没有公开数据，但其 1.5B 版本开箱即用的得分约为 73%——尽管它在 int8 内存中占用约 1.5GB 的空间，是 FunctionGemma（288MB）的 5 倍。
In September 2025, she launched a business offering support to others in Guernsey because of her experience.
，推荐阅读同城约会获取更多信息
// Nothing stops you from doing this
Kalshi, one of several online prediction markets that have exploded in popularity in the last few years, has suspended one of YouTube MrBeast's video editors for insider trading, NPR reports. Besides being suspended from the platform for two years, Kalshi says the editor will also be required to pay a financial penalty that's five times his initial trade size.