RecursiveSI Flint Benchmark: unsloth/flint-GGUF flint-UD-Q3_K_XL.gguf
英文标题：RecursiveSI Flint Benchmark: unsloth/flint-GGUF flint-UD-Q3_K_XL.gguf
Unsloth Flint Q3 Results Performed quite poorly on benchmark using 3 diff harnesses: OMP: Completed all 12 tasks; 71.2/100 (only 7 solved). AtomicAgent: Completed all 12 tasks; 39.2/100 (only 3 solved). Hermes: Stopped during Task 7 after it performed needless broad searches across the filesy
打开原文相关 4 · 机会 0
中文
Unsloth Flint Q3 Results Performed quite poorly on benchmark using 3 diff harnesses: OMP: Completed all 12 tasks; 71.2/100 (only 7 solved). AtomicAgent: Completed all 12 tasks; 39.2/100 (only 3 solved). Hermes: Stopped during Task 7 after it performed needless broad searches across the filesy
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
Unsloth Flint Q3 Results Performed quite poorly on benchmark using 3 diff harnesses: OMP: Completed all 12 tasks; 71.2/100 (only 7 solved). AtomicAgent: Completed all 12 tasks; 39.2/100 (only 3 solved). Hermes: Stopped during Task 7 after it performed needless broad searches across the filesy