跳到正文
RCreddit.com·
暂不在当前实时榜单

A Strata fork for IBM AC922 running Qwen3.8-FN UD-Q4_K_XL is doing up to 7,357 tk/s prefill and 113 tk/s decode

AI 摘要

A developer forked Strata, making significant changes to Opus 5.5, to run it on an IBM AC922. This 2018-era IBM AC922, equipped with two POWER9 20-core SMT4 CPUs and 4 or 6 NVIDIA Tesla V100 SXM2 GPUs, achieved impressive performance. Running Qwen3.8-FN UD-Q4_K_XL, the system reached up to 7,357 tk/s for prefill and 113 tk/s for decode, leveraging the 150GB/s CPU-GPU bandwidth and unified memory access.

为什么是这条

This report details a Strata fork that achieves 7,357 tk/s prefill and 113 tk/s decode on an IBM AC922, unlike typical benchmarks on newer hardware.

时间与来源

时间显示为 UTC

显示时区:UTC

本地时区尚不可用,暂时显示 UTC。

收录当时偏移:UTC+02026年10月4日 21:00 UTC

收录
2026年10月4日 21:00
来源类型
开发者社区

本站未收录正文。

前往源站阅读 →
来源·reddit.com