Round score
0
0.00%
Prompts scored
0 / 0
median_e2e_speedup
Below tolerance
0
answered too little to score
Zeroed total
0
scored 0.0 without measuring
Temperature range
Request interval
Thinking
Context limit
The same inputs and output limits apply to every engine. Actual concurrency depends on response duration and engine scheduling.
The round score is the scoring rule applied to the prompts below. Both engines are compared at the same output token count, so a candidate that stops early cannot buy speed by answering less: a prompt under the tolerance bar scores nothing and shows its gate reason.
Reference engine
The baseline is what candidates are measured against, so it has no speedup of its own. Its SLA replay is below.
32 requests
| Request | TTFT | Total | Tokens | Slowest gap | Stream |
|---|---|---|---|---|---|
| hf-000 | 238.25ms | 91.759s | 4778 | 330.72ms | — |
| hf-001 | 251.02ms | 81.598s | 4178 | 318.26ms | — |
| hf-002 | 248.87ms | 95.782s |
Actual settings for each measured request. Each prompt keeps its temperature across repetitions and engines. Generation uses seed 0 throughout qualification, warmups and measured repetitions; fixed settings do not guarantee identical output text.
| Prompt | Repetition | Temperature | Top p | Seed |
|---|---|---|---|---|
| hf-003 | 1 | 0.997721 | 1 | 0 |
| hf-008 | 1 | 0.604643 | 1 | 0 |
| hf-014 | 1 | 0.377566 | 1 | 0 |
| hf-022 | 1 | 0.253973 | 1 | 0 |
| hf-010 | 1 | 0.624855 | 1 |
| 5120 |
| 317.70ms |
| — |
| hf-003 | 246.82ms | 69.900s | 3579 | 316.60ms | — |
| hf-004 | 245.16ms | 87.716s | 4512 | 316.14ms | — |
| hf-005 | 259.31ms | 96.280s | 5120 | 300.45ms | — |
| hf-006 | 257.73ms | 94.312s | 4960 | 299.96ms | — |
| hf-007 | 254.97ms | 86.649s | 4447 | 299.87ms | — |
| hf-008 | 315.26ms | 79.712s | 4077 | 204.54ms | — |
| hf-009 | 310.19ms | 90.549s | 4733 | 241.97ms | — |
| hf-010 | 338.97ms | 86.544s | 4441 | 211.26ms | — |
| hf-011 | 328.57ms | 95.764s | 5120 | 219.90ms | — |
| hf-012 | 335.09ms | 96.261s | 5120 | 211.09ms | — |
| hf-013 | 360.69ms | 96.264s | 5120 | 183.14ms | — |
| hf-014 | 384.14ms | 91.988s | 4803 | 123.17ms | — |
| hf-015 | 356.48ms | 96.260s | 5120 | 183.17ms | — |
| hf-016 | 406.63ms | 84.497s | 4354 | 132.71ms | — |
| hf-017 | 419.78ms | 87.251s | 4505 | 84.24ms | — |
| hf-018 | 417.48ms | 87.684s | 4532 | 83.17ms | — |
| hf-019 | 429.23ms | 84.538s | 4323 | 103.06ms | — |
| hf-020 | 434.22ms | 82.587s | 4238 | 97.49ms | — |
| hf-021 | 413.61ms | 78.770s | 4030 | 81.25ms | — |
| hf-022 | 425.76ms | 94.927s | 5012 | 100.61ms | — |
| hf-023 | 423.92ms | 96.244s | 5120 | 100.61ms | — |
| hf-024 | 421.86ms | 96.250s | 5120 | 100.15ms | — |
| hf-025 | 458.13ms | 95.736s | 5120 | 63.52ms | — |
| hf-026 | 456.32ms | 81.202s | 4160 | 63.57ms | — |
| hf-027 | 448.51ms | 89.431s | 4621 | 68.05ms | — |
| hf-028 | 491.66ms | 89.932s | 4692 | 33.23ms | — |
| hf-029 | 490.24ms | 83.327s | 4284 | 33.59ms | — |
| hf-030 | 485.96ms | 96.230s | 5120 | 33.29ms | — |
| hf-031 | 486.70ms | 82.954s | 4241 | 33.27ms | — |
| 0 |
| hf-026 | 1 | 0.895267 | 1 | 0 |
| hf-001 | 1 | 0.853685 | 1 | 0 |
| hf-027 | 1 | 0.81344 | 1 | 0 |
| hf-000 | 1 | 0.693931 | 1 | 0 |
| hf-031 | 1 | 0.544279 | 1 | 0 |
| hf-020 | 1 | 0.415697 | 1 | 0 |
| hf-029 | 1 | 0.384713 | 1 | 0 |
| hf-016 | 1 | 0.494597 | 1 | 0 |
| hf-019 | 1 | 0.40575 | 1 | 0 |
| hf-017 | 1 | 0.286229 | 1 | 0 |
| hf-028 | 1 | 0.805708 | 1 | 0 |
| hf-009 | 1 | 0.921266 | 1 | 0 |
| hf-012 | 1 | 0.837466 | 1 | 0 |
| hf-018 | 1 | 0.565021 | 1 | 0 |
| hf-006 | 1 | 0.196355 | 1 | 0 |
| hf-023 | 1 | 0.452587 | 1 | 0 |
| hf-030 | 1 | 0.131211 | 1 | 0 |
| hf-021 | 1 | 0.277881 | 1 | 0 |
| hf-011 | 1 | 0.220758 | 1 | 0 |
| hf-004 | 1 | 0.593518 | 1 | 0 |
| hf-013 | 1 | 0.733351 | 1 | 0 |
| hf-024 | 1 | 0.733266 | 1 | 0 |
| hf-005 | 1 | 0.231691 | 1 | 0 |
| hf-002 | 1 | 0.998803 | 1 | 0 |
| hf-025 | 1 | 0.104149 | 1 | 0 |
| hf-007 | 1 | 0.891091 | 1 | 0 |
| hf-015 | 1 | 0.24944 | 1 | 0 |
| hf-003 | 2 | 0.997721 | 1 | 0 |
| hf-021 | 2 | 0.277881 | 1 | 0 |
| hf-001 | 2 | 0.853685 | 1 | 0 |
| hf-026 | 2 | 0.895267 | 1 | 0 |
| hf-029 | 2 | 0.384713 | 1 | 0 |
| hf-018 | 2 | 0.565021 | 1 | 0 |
| hf-031 | 2 | 0.544279 | 1 | 0 |
| hf-016 | 2 | 0.494597 | 1 | 0 |
| hf-019 | 2 | 0.40575 | 1 | 0 |
| hf-017 | 2 | 0.286229 | 1 | 0 |
| hf-010 | 2 | 0.624855 | 1 | 0 |
| hf-007 | 2 | 0.891091 | 1 | 0 |
| hf-004 | 2 | 0.593518 | 1 | 0 |
| hf-009 | 2 | 0.921266 | 1 | 0 |
| hf-027 | 2 | 0.81344 | 1 | 0 |
| hf-020 | 2 | 0.415697 | 1 | 0 |
| hf-000 | 2 | 0.693931 | 1 | 0 |
| hf-025 | 2 | 0.104149 | 1 | 0 |
| hf-008 | 2 | 0.604643 | 1 | 0 |
| hf-006 | 2 | 0.196355 | 1 | 0 |
| hf-011 | 2 | 0.220758 | 1 | 0 |
| hf-022 | 2 | 0.253973 | 1 | 0 |
| hf-014 | 2 | 0.377566 | 1 | 0 |
| hf-012 | 2 | 0.837466 | 1 | 0 |
| hf-023 | 2 | 0.452587 | 1 | 0 |
| hf-002 | 2 | 0.998803 | 1 | 0 |
| hf-015 | 2 | 0.24944 | 1 | 0 |
| hf-030 | 2 | 0.131211 | 1 | 0 |
| hf-024 | 2 | 0.733266 | 1 | 0 |
| hf-005 | 2 | 0.231691 | 1 | 0 |
| hf-028 | 2 | 0.805708 | 1 | 0 |
| hf-013 | 2 | 0.733351 | 1 | 0 |
| hf-020 | 3 | 0.415697 | 1 | 0 |
| hf-004 | 3 | 0.593518 | 1 | 0 |
| hf-028 | 3 | 0.805708 | 1 | 0 |
| hf-021 | 3 | 0.277881 | 1 | 0 |
| hf-008 | 3 | 0.604643 | 1 | 0 |
| hf-002 | 3 | 0.998803 | 1 | 0 |
| hf-001 | 3 | 0.853685 | 1 | 0 |
| hf-007 | 3 | 0.891091 | 1 | 0 |
| hf-019 | 3 | 0.40575 | 1 | 0 |
| hf-017 | 3 | 0.286229 | 1 | 0 |
| hf-010 | 3 | 0.624855 | 1 | 0 |
| hf-018 | 3 | 0.565021 | 1 | 0 |
| hf-029 | 3 | 0.384713 | 1 | 0 |
| hf-031 | 3 | 0.544279 | 1 | 0 |
| hf-006 | 3 | 0.196355 | 1 | 0 |
| hf-014 | 3 | 0.377566 | 1 | 0 |
| hf-027 | 3 | 0.81344 | 1 | 0 |
| hf-000 | 3 | 0.693931 | 1 | 0 |
| hf-015 | 3 | 0.24944 | 1 | 0 |
| hf-030 | 3 | 0.131211 | 1 | 0 |
| hf-011 | 3 | 0.220758 | 1 | 0 |
| hf-012 | 3 | 0.837466 | 1 | 0 |
| hf-025 | 3 | 0.104149 | 1 | 0 |
| hf-026 | 3 | 0.895267 | 1 | 0 |
| hf-023 | 3 | 0.452587 | 1 | 0 |
| hf-016 | 3 | 0.494597 | 1 | 0 |
| hf-024 | 3 | 0.733266 | 1 | 0 |
| hf-009 | 3 | 0.921266 | 1 | 0 |
| hf-005 | 3 | 0.231691 | 1 | 0 |
| hf-022 | 3 | 0.253973 | 1 | 0 |
| hf-013 | 3 | 0.733351 | 1 | 0 |
| hf-003 | 3 | 0.997721 | 1 | 0 |
role
evidence
requests per s
sla goodput ratio
output tokens per s
p99 e2e ms rel range
p99 itl ms rel range
p99 ttft ms rel range
Image digest