AI 导读
端到端生成时间,以 1,024 token 输入后生成 256 个输出 token 的总耗时衡量,在我们于 iPhone 17 Pro 上测试的模型间相差 30 倍,从 LFM2.5-230M 的 0.9s 到 Falcon-H1R-7B 的 26.7s。并列最高分的两个模型分别为 8.0s(LFM2.5-2.6B)和 21.4s(Nanbeige4.2-3B)
正文
End-to-End Generation Time, measured as the total time to generate 256 output tokens after a 1,024 token input, spans 30x across the models we tested on an iPhone 17 Pro, from 0.9s for LFM2.5-230M to 26.7s for Falcon-H1R-7B. The two models tied for the top score sit at 8.0s (LFM2.5-2.6B) and 21.4s (Nanbeige4.2-3B)
来源:@ArtificialAnlys · x.com