Round score
0
weighted tier speedup after failure cap and deduction
Eligible requests
—
trusted baseline exclusions removed
Failed eligible requests
—
each request counted once
Raw weighted speedup
—
before failure cap and deduction
Temperature range
Request concurrency
Tier scheduling
Output policy
Thinking
Context limit
The same eligible inputs and output ceilings apply to every engine. Candidates must emit at least 90% of baseline tokens per request; natural output lengths can affect tier completion time. Slots refill until the group drains and stay occupied through protocol completion. Baseline exclusions and final drain can reduce actual concurrency below the configured cap.
The round score uses weighted completion times for all eligible work in each tier. Per-request speedups below are diagnostics, not score contributions. Only trusted baseline exclusions are removed; candidate failures remain in the eligible count and incur the configured failure penalty.
Reference engine
The baseline is what candidates are measured against, so it has no speedup of its own. Its SLA replay is below.
31 requests
| Request | TTFT | Total | Tokens | Slowest gap | Stream |
|---|---|---|---|---|---|
| hf-000 | 631.99ms | 69.366s | 3817 | 111.58ms | 1792 gaps for 3817 tokens |
| hf-001 | 716.56ms | 76.140s | 4166 | 41.31ms | 1946 gaps for 4166 tokens |
| hf-002 | 774.23ms |
Occupied client slots include protocol completion and final drain. Mean occupancy is weighted by elapsed time; it can be below the requested cap.
| Repetition | Group start | Requested | Effective cap | Peak | Mean |
|---|---|---|---|---|---|
| 1 | 0.000s | C32 | 32 | 32 | 23.70 |
| 2 | 0.000s | C32 | 32 | 32 | 25.71 |
| 3 | 0.000s | C32 | 32 | 32 | 21.90 |
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 |
|---|
| 63.643s |
| 3357 |
| 44.40ms |
| 1661 gaps for 3357 tokens |
| hf-003 | 719.08ms | 97.779s | 5120 | 43.94ms | 2668 gaps for 5120 tokens |
| hf-004 | 716.71ms | 70.984s | 3463 | 41.27ms | 1839 gaps for 3463 tokens |
| hf-005 | 741.12ms | 98.742s | 5120 | 43.98ms | 2828 gaps for 5120 tokens |
| hf-006 | 710.49ms | 91.777s | 5120 | 41.30ms | 2393 gaps for 5120 tokens |
| hf-007 | 705.13ms | 69.932s | 3655 | 42.42ms | 1781 gaps for 3655 tokens |
| hf-008 | 769.44ms | 99.796s | 5120 | 46.39ms | 2872 gaps for 5120 tokens |
| hf-009 | 710.82ms | 92.858s | 5120 | 41.57ms | 2427 gaps for 5120 tokens |
| hf-010 | 771.49ms | 91.181s | 4494 | 44.14ms | 2537 gaps for 4494 tokens |
| hf-011 | 721.19ms | 93.357s | 4993 | 43.33ms | 2508 gaps for 4993 tokens |
| hf-012 | 713.43ms | 67.508s | 3273 | 42.32ms | 1717 gaps for 3273 tokens |
| hf-013 | 705.26ms | 99.254s | 4612 | 42.53ms | 2667 gaps for 4612 tokens |
| hf-014 | 711.79ms | 71.961s | 3519 | 43.67ms | 1835 gaps for 3519 tokens |
| hf-015 | 713.71ms | 106.807s | 4920 | 43.39ms | 3103 gaps for 4920 tokens |
| hf-016 | 711.14ms | 70.401s | 4166 | 42.54ms | 1822 gaps for 4166 tokens |
| hf-018 | 704.48ms | 55.376s | 2913 | 42.50ms | 1416 gaps for 2913 tokens |
| hf-019 | 709.23ms | 108.749s | 5120 | 43.63ms | 3229 gaps for 5120 tokens |
| hf-020 | 712.12ms | 92.448s | 5120 | 42.57ms | 2479 gaps for 5120 tokens |
| hf-021 | 704.19ms | 99.562s | 5120 | 42.54ms | 2680 gaps for 5120 tokens |
| hf-022 | 707.93ms | 97.165s | 5120 | 42.35ms | 2645 gaps for 5120 tokens |
| hf-023 | 750.27ms | 96.383s | 5120 | 44.55ms | 2737 gaps for 5120 tokens |
| hf-024 | 707.93ms | 94.570s | 4812 | 42.53ms | 2485 gaps for 4812 tokens |
| hf-025 | 701.57ms | 55.161s | 2930 | 41.70ms | 1403 gaps for 2930 tokens |
| hf-026 | 707.66ms | 103.742s | 5120 | 43.06ms | 2939 gaps for 5120 tokens |
| hf-027 | 755.47ms | 53.468s | 2895 | 45.59ms | 1378 gaps for 2895 tokens |
| hf-028 | 752.26ms | 68.824s | 3355 | 44.66ms | 1815 gaps for 3355 tokens |
| hf-029 | 758.40ms | 56.247s | 3062 | 44.64ms | 1454 gaps for 3062 tokens |
| hf-030 | 697.12ms | 94.842s | 5120 | 42.30ms | 2496 gaps for 5120 tokens |
| hf-031 | 702.54ms | 87.243s | 4660 | 42.41ms | 2259 gaps for 4660 tokens |
| Top p |
|---|
| Seed |
|---|
| hf-027 | 1 | 0.549717 | 1 | 0 |
| hf-014 | 1 | 0.393375 | 1 | 0 |
| hf-013 | 1 | 0.233417 | 1 | 0 |
| hf-029 | 1 | 0.94331 | 1 | 0 |
| hf-018 | 1 | 0.566662 | 1 | 0 |
| hf-012 | 1 | 0.617887 | 1 | 0 |
| hf-010 | 1 | 0.347258 | 1 | 0 |
| hf-028 | 1 | 0.489132 | 1 | 0 |
| hf-017 | 1 | 0.704009 | 1 | 0 |
| hf-000 | 1 | 0.907409 | 1 | 0 |
| hf-031 | 1 | 0.722809 | 1 | 0 |
| hf-016 | 1 | 0.983149 | 1 | 0 |
| hf-004 | 1 | 0.27719 | 1 | 0 |
| hf-005 | 1 | 0.382033 | 1 | 0 |
| hf-002 | 1 | 0.6371 | 1 | 0 |
| hf-008 | 1 | 0.36576 | 1 | 0 |
| hf-020 | 1 | 0.971895 | 1 | 0 |
| hf-011 | 1 | 0.750363 | 1 | 0 |
| hf-025 | 1 | 0.722464 | 1 | 0 |
| hf-009 | 1 | 0.932255 | 1 | 0 |
| hf-007 | 1 | 0.912541 | 1 | 0 |
| hf-024 | 1 | 0.649567 | 1 | 0 |
| hf-022 | 1 | 0.38995 | 1 | 0 |
| hf-003 | 1 | 0.577771 | 1 | 0 |
| hf-006 | 1 | 0.949 | 1 | 0 |
| hf-021 | 1 | 0.488697 | 1 | 0 |
| hf-023 | 1 | 0.405127 | 1 | 0 |
| hf-026 | 1 | 0.241993 | 1 | 0 |
| hf-015 | 1 | 0.139166 | 1 | 0 |
| hf-030 | 1 | 0.452432 | 1 | 0 |
| hf-019 | 1 | 0.104092 | 1 | 0 |
| hf-001 | 1 | 0.829672 | 1 | 0 |
| hf-016 | 2 | 0.983149 | 1 | 0 |
| hf-025 | 2 | 0.722464 | 1 | 0 |
| hf-002 | 2 | 0.6371 | 1 | 0 |
| hf-012 | 2 | 0.617887 | 1 | 0 |
| hf-007 | 2 | 0.912541 | 1 | 0 |
| hf-004 | 2 | 0.27719 | 1 | 0 |
| hf-014 | 2 | 0.393375 | 1 | 0 |
| hf-001 | 2 | 0.829672 | 1 | 0 |
| hf-029 | 2 | 0.94331 | 1 | 0 |
| hf-023 | 2 | 0.405127 | 1 | 0 |
| hf-027 | 2 | 0.549717 | 1 | 0 |
| hf-031 | 2 | 0.722809 | 1 | 0 |
| hf-011 | 2 | 0.750363 | 1 | 0 |
| hf-006 | 2 | 0.949 | 1 | 0 |
| hf-009 | 2 | 0.932255 | 1 | 0 |
| hf-028 | 2 | 0.489132 | 1 | 0 |
| hf-017 | 2 | 0.704009 | 1 | 0 |
| hf-018 | 2 | 0.566662 | 1 | 0 |
| hf-020 | 2 | 0.971895 | 1 | 0 |
| hf-003 | 2 | 0.577771 | 1 | 0 |
| hf-000 | 2 | 0.907409 | 1 | 0 |
| hf-024 | 2 | 0.649567 | 1 | 0 |
| hf-030 | 2 | 0.452432 | 1 | 0 |
| hf-013 | 2 | 0.233417 | 1 | 0 |
| hf-021 | 2 | 0.488697 | 1 | 0 |
| hf-022 | 2 | 0.38995 | 1 | 0 |
| hf-008 | 2 | 0.36576 | 1 | 0 |
| hf-005 | 2 | 0.382033 | 1 | 0 |
| hf-010 | 2 | 0.347258 | 1 | 0 |
| hf-026 | 2 | 0.241993 | 1 | 0 |
| hf-019 | 2 | 0.104092 | 1 | 0 |
| hf-015 | 2 | 0.139166 | 1 | 0 |
| hf-017 | 3 | 0.704009 | 1 | 0 |
| hf-025 | 3 | 0.722464 | 1 | 0 |
| hf-018 | 3 | 0.566662 | 1 | 0 |
| hf-020 | 3 | 0.971895 | 1 | 0 |
| hf-030 | 3 | 0.452432 | 1 | 0 |
| hf-001 | 3 | 0.829672 | 1 | 0 |
| hf-027 | 3 | 0.549717 | 1 | 0 |
| hf-007 | 3 | 0.912541 | 1 | 0 |
| hf-029 | 3 | 0.94331 | 1 | 0 |
| hf-022 | 3 | 0.38995 | 1 | 0 |
| hf-000 | 3 | 0.907409 | 1 | 0 |
| hf-002 | 3 | 0.6371 | 1 | 0 |
| hf-006 | 3 | 0.949 | 1 | 0 |
| hf-028 | 3 | 0.489132 | 1 | 0 |
| hf-016 | 3 | 0.983149 | 1 | 0 |
| hf-024 | 3 | 0.649567 | 1 | 0 |
| hf-012 | 3 | 0.617887 | 1 | 0 |
| hf-031 | 3 | 0.722809 | 1 | 0 |
| hf-009 | 3 | 0.932255 | 1 | 0 |
| hf-010 | 3 | 0.347258 | 1 | 0 |
| hf-021 | 3 | 0.488697 | 1 | 0 |
| hf-011 | 3 | 0.750363 | 1 | 0 |
| hf-023 | 3 | 0.405127 | 1 | 0 |
| hf-005 | 3 | 0.382033 | 1 | 0 |
| hf-008 | 3 | 0.36576 | 1 | 0 |
| hf-004 | 3 | 0.27719 | 1 | 0 |
| hf-014 | 3 | 0.393375 | 1 | 0 |
| hf-013 | 3 | 0.233417 | 1 | 0 |
| hf-026 | 3 | 0.241993 | 1 | 0 |
| hf-015 | 3 | 0.139166 | 1 | 0 |
| hf-019 | 3 | 0.104092 | 1 | 0 |
| hf-003 | 3 | 0.577771 | 1 | 0 |
role
evidence
request concurrency
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
tier completion max rel range
Image digest