DeepSeek says V4.1-Flash-1017 was trained for their own harness, so I tested it. Same model (deepseek-flash API, thinking high), same 34 coding/shell/git tasks, 2 runs each, interleaved. Quality tied (407/408 passed). But DeepSeek Harness took 1.63x the time and 1.91x the output tokens vs minimal Pi. Swapping Pi's tool schemas to DSH's: 1.34x slower. Switching Pi to DeepSeek's Messages API: 0.81x,...

1. 

   消息推送

   DeepSeek says V4.1-Flash-1017 was trained for their own harness, so I tested it. Same model (deepseek-flash API, thinking high), same 34 coding/shell/git tasks, 2 runs each, interleaved. Quality tied (407/408 passed). But DeepSeek Harness took 1.63x the time and 1.91x the output tokens vs minimal Pi. Swapping Pi's tool schemas to DSH's: 1.34x slower. Switching Pi to DeepSeek's Messages API: 0.81x,...

   5天前

   DeepSeek says V4.1-Flash-1017 was trained for their own harness, so I tested it. Same model (deepseek-flash API, thinking high), same 34 coding/shell/git tasks, 2 runs each, interleaved. Quality tied (407/408 passed). But DeepSeek Harness took 1.63x the time and 1.91x the output tokens vs minimal Pi. Swapping Pi's tool schemas to DSH's: 1.34x slower. Switching Pi to DeepSeek's Messages API: 0.81x,...
   
   🌐 链接: https://x.com/zengjiajun_eth/status/2111456956544115165
   🏷️ 分组: X｜为你推荐
   🕒 时间: 2026-10-17 21:59:02

   X (formerly Twitter)

   

   Zeng Jiajun 曾嘉俊 (@zengjiajun_eth) on X

   DeepSeek says V4.1-Flash-1017 was trained for their own harness, so I tested it.
   
   Same model (deepseek-flash API, thinking high), same 34 coding/shell/git tasks, 2 runs each, interleaved.
   
   Quality…

   5天前