Strata - 512K YaRN tested! · @indesjyo · TwKit

@indesjyo

The result I measured Qwen3.8-Flash-Next IQ3_S on one RTX 5090 32 GB, Ryzen 7 9700X, 93.7 GB RAM, Windows 11, Strata 0.1.45. A single cached chat at 64K, 128K, and 256K returns the first token in well

曝光 286 · 评论 0 · 点赞 9 · 书签 4