sglang/ GB200 perf

dsr1-fp4-1k1k-mid-curve

concurrency 2,048 · 3 days ago

cron
passed
re-run
Commit

[diffusion] fix: validate openai sampling dimensions (#28791)

mickqian·3 days ago
PR #28791 · Validate OpenAI sampling dimensions
Run
GitHub Actions27892080060-1
Slurm job5327
GPUs48·prefill 16 / decode 32
ISL / OSL1024 / 1024

Metrics

28 captured

best_of

1.00

burstiness

1.00

completed

20,480

duration

216

max_concurrency

2,048

mean_e2el_ms

20,799ms

mean_itl_ms

1,012ms

mean_tpot_ms

20.64ms

mean_ttft_ms

1,781ms

median_e2el_ms

20,354ms

median_itl_ms

1,047ms

median_tpot_ms

21.04ms

median_ttft_ms

830ms

num_prompts

20,480

output_throughput

87,616tok/s

p99_e2el_ms

32,991ms

p99_itl_ms

1,637ms

p99_tpot_ms

22.18ms

p99_ttft_ms

13,639ms

peak_output_tokens_per_s

104,182s

request_throughput

95.00tok/s

std_e2el_ms

3,011ms

std_itl_ms

217ms

std_tpot_ms

1.37ms

std_ttft_ms

2,822ms

total_input_tokens

18,880,692

total_output_tokens

18,888,974

total_token_throughput

175,193tok/s