dsr1-fp8-1k1k-max-tpt

concurrency 2,048 · 9 days ago

cron

passed

re-run

Commit

SHA0417951

[Bug Fix] Validate tokenizer-dependent features with skip_tokenizer_init (#27882)

rllin·9 days ago

PR #27882 · [Bug Fix] Validate tokenizer-dependent features with skip_tokenizer_init

Run

GitHub Actions27522040539-1

Slurm job5309

GPUs48·prefill 16 / decode 32

ISL / OSL1024 / 1024

Logscron/27522040539-1/1k1k/dsr1-fp8-1k1k-max-tpt/2026-06-15/5309/sa-bench_isl_1024_osl_1024/

Metrics

28 captured

best_of

1.00

burstiness

1.00

completed

20,480

duration

284

max_concurrency

2,048

mean_e2el_ms

27,240ms

mean_itl_ms

1,344ms

mean_tpot_ms

27.52ms

mean_ttft_ms

1,884ms

median_e2el_ms

26,830ms

median_itl_ms

1,379ms

median_tpot_ms

27.62ms

median_ttft_ms

1,014ms

num_prompts

20,480

output_throughput

66,523tok/s

p99_e2el_ms

37,772ms

p99_itl_ms

1,932ms

p99_tpot_ms

28.89ms

p99_ttft_ms

11,936ms

peak_output_tokens_per_s

79,542s

request_throughput

72.13tok/s

std_e2el_ms

2,924ms

std_itl_ms

233ms

std_tpot_ms

0.96ms

std_ttft_ms

2,564ms

total_input_tokens

18,880,692

total_output_tokens

18,888,974

total_token_throughput

133,017tok/s