sglang/ GB200 perf

dsr1-fp4-1k1k-mid-curve

concurrency 2,048 · 12 days ago

cron
passed
re-run
Commit

add lfm2.5 to new cookbook. (#27409)

vincentzed·12 days ago
PR #27409 · add lfm2.5 to new cookbook.
Run
GitHub Actions27392321842-1
Slurm job5299
GPUs48·prefill 16 / decode 32
ISL / OSL1024 / 1024

Metrics

28 captured

best_of

1.00

burstiness

1.00

completed

20,480

duration

208

max_concurrency

2,048

mean_e2el_ms

20,075ms

mean_itl_ms

978ms

mean_tpot_ms

19.94ms

mean_ttft_ms

1,698ms

median_e2el_ms

19,694ms

median_itl_ms

1,007ms

median_tpot_ms

20.29ms

median_ttft_ms

823ms

num_prompts

20,480

output_throughput

90,740tok/s

p99_e2el_ms

30,757ms

p99_itl_ms

1,548ms

p99_tpot_ms

21.30ms

p99_ttft_ms

11,877ms

peak_output_tokens_per_s

105,064s

request_throughput

98.38tok/s

std_e2el_ms

2,734ms

std_itl_ms

194ms

std_tpot_ms

1.30ms

std_ttft_ms

2,558ms

total_input_tokens

18,880,692

total_output_tokens

18,888,974

total_token_throughput

181,440tok/s