dsr1-fp4-1k1k-mid-curve

best_of

1.00

burstiness

1.00

completed

20,480

duration

211

max_concurrency

2,048

mean_e2el_ms

20,319ms

mean_itl_ms

993ms

mean_tpot_ms

20.25ms

mean_ttft_ms

1,660ms

median_e2el_ms

19,944ms

median_itl_ms

1,026ms

median_tpot_ms

20.65ms

median_ttft_ms

823ms

num_prompts

20,480

output_throughput

89,559tok/s

p99_e2el_ms

31,282ms

p99_itl_ms

1,531ms

p99_tpot_ms

21.49ms

p99_ttft_ms

12,209ms

peak_output_tokens_per_s

104,207s

request_throughput

97.10tok/s

std_e2el_ms

2,748ms

std_itl_ms

202ms

std_tpot_ms

1.34ms

std_ttft_ms

2,534ms

total_input_tokens

18,880,692

total_output_tokens

18,888,974

total_token_throughput

179,078tok/s