sglang/ GB200 perf

dsr1-fp8-1k1k-max-tpt

concurrency 2,048 · 5 days ago

manual
passed
re-run
Commit

fix dependencies

csahithi·5 days ago
Run
GitHub Actions24591826053-2
Slurm job4667
GPUs48·prefill 16 / decode 32
ISL / OSL1024 / 1024

Metrics

27 captured

best_of

1.00

burstiness

1.00

completed

20,480

duration

386

max_concurrency

2,048

mean_e2el_ms

36,450ms

mean_itl_ms

868ms

mean_tpot_ms

17.83ms

mean_ttft_ms

20,028ms

median_e2el_ms

36,300ms

median_itl_ms

883ms

median_tpot_ms

17.69ms

median_ttft_ms

19,800ms

num_prompts

20,480

output_throughput

48,897tok/s

p99_e2el_ms

52,196ms

p99_itl_ms

1,220ms

p99_tpot_ms

18.32ms

p99_ttft_ms

35,613ms

request_throughput

53.02tok/s

std_e2el_ms

3,731ms

std_itl_ms

142ms

std_tpot_ms

0.22ms

std_ttft_ms

3,559ms

total_input_tokens

18,880,692

total_output_tokens

18,888,974

total_token_throughput

97,772tok/s