sglang/ GB200 perf

dsr1-fp8-1k1k-max-tpt

concurrency 2,048 · 21 hours ago

cron
passed
re-run
Commit

Vectorize _create_custom_4d_mask in CustomQwen2Decoder (#27527)

ckvermaAI·22 hours ago
PR #27527 · Vectorize _create_custom_4d_mask in CustomQwen2Decoder
Run
GitHub Actions27999739491-1
Slurm job5333
GPUs48·prefill 16 / decode 32
ISL / OSL1024 / 1024

Metrics

28 captured

best_of

1.00

burstiness

1.00

completed

20,480

duration

270

max_concurrency

2,048

mean_e2el_ms

25,854ms

mean_itl_ms

1,268ms

mean_tpot_ms

25.96ms

mean_ttft_ms

1,936ms

median_e2el_ms

25,326ms

median_itl_ms

1,296ms

median_tpot_ms

25.99ms

median_ttft_ms

1,018ms

num_prompts

20,480

output_throughput

69,906tok/s

p99_e2el_ms

37,403ms

p99_itl_ms

1,762ms

p99_tpot_ms

26.83ms

p99_ttft_ms

12,376ms

peak_output_tokens_per_s

81,298s

request_throughput

75.79tok/s

std_e2el_ms

3,040ms

std_itl_ms

210ms

std_tpot_ms

0.70ms

std_ttft_ms

2,668ms

total_input_tokens

18,880,692

total_output_tokens

18,888,974

total_token_throughput

139,782tok/s