For Qwen3-8B using the largest fitting batch, calibrated TP=2 reached 2.16× on AIME25 and 2.07× on LB[16k-18k], above the table’s TP=1 runs. The timing model achieved 5–11% MAPE when refit at either TP degree; reusing TP=1 coefficients at TP=2 raised MAPE to 35–50%, left AIME speedup benign at 2.21×, but reduced LB speedup to 1.51×.

报告指标与观测值

Assessment（0）

这条 Claim 暂无已发布的不可变 Assessment。

关联运行（0）

这条 Claim 暂无关联执行。

每行是一项测量，分别显示论文报告、最近可用实测和累计证据状态。
指标与条件	论文报告	最近可用实测	证据判断
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1	30 requests	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1	1,353 output token/s	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1	2,553 output token/s	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1	1.89 × AutoRegressive	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2	64 requests	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2	2,494 output token/s	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2	5,392 output token/s	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2	2.16 × AutoRegressive	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale	64 requests	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale	2,494 output token/s	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale	5,509 output token/s	—	尚未评估
model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale	2.21 × AutoRegressive	—	尚未评估

指标与条件

论文报告

最近可用实测

证据判断

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1

30 requests

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1

1,353 output token/s

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1

2,553 output token/s

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 1

1.89 × AutoRegressive

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2

64 requests

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2

2,494 output token/s

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2

5,392 output token/s

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: fitted for this TP degree · tensorParallelism: 2

2.16 × AutoRegressive

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale

64 requests

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale

2,494 output token/s

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale

5,509 output token/s

—

尚未评估

model: Qwen3-8B · dataset: AIME25 · calibration: TP=1 coefficients reused at TP=2 · tensorParallelism: 2 stale

2.21 × AutoRegressive

—

尚未评估

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果