Round score
0.1194
weighted tier speedup after failure cap and deduction
Eligible requests
31
trusted baseline exclusions removed
Failed eligible requests
13
each request counted once
Raw weighted speedup
+16.14%
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.
Raw weighted speedup
Eligible speedup
Request failures cap speed credit at zero: min(0.161377, 0) = 0.161377. Negative speedups remain negative.
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.
31 prompts
| Prompt | Diagnostic speedup | Baseline | Candidate | Aligned output | Diagnostic / failure |
|---|---|---|---|---|---|
| hf-027 | — | — | — | 2895 | Per-token timing unavailable (not a request failure) |
| hf-014 | — | — | — | 3519 | Per-token timing unavailable (not a request failure) |
| hf-013 | — | — | — |
31 requests
| Request | TTFT | Total | Tokens | Slowest gap | Stream |
|---|---|---|---|---|---|
| hf-000 | 628.35ms | 38.673s | 2347 | 108.81ms | 1004 gaps for 2347 tokens |
| hf-001 | 710.28ms | 73.820s | 5120 | 39.54ms | 2236 gaps for 5120 tokens |
| hf-002 | 712.71ms |
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 | 31 | 31 | 25.56 |
| 2 | 0.000s | C32 | 31 | 31 | 22.98 |
| 3 | 0.000s | C32 | 31 | 31 | 25.28 |
Each latency-median response is compared with the lowest valid opening-baseline ratio for the same prompt. Drops above the recorded limit fail; absolute checks also apply. These checks detect repetition, not overall writing quality.
| Prompt | Candidate char-16 |
|---|
Failed requests
Deduction
Final score
Each duration is the median across repetitions, from the group start until the tier's last request finishes. This includes client queueing. Weighted speedup = sum of weight × (1 − candidate / baseline).
| Input tier | Eligible requests | Weight | Baseline | Candidate | Speedup | Contribution |
|---|---|---|---|---|---|---|
| 2k | 8 | 25% | 110.927s | 89.526s | +19.29% | +4.82% |
| 4k | 7 | 25% | 100.129s | 87.374s | +12.74% | +3.18% |
| 8k | 8 | 25% | 106.824s | 93.080s | +12.87% | +3.22% |
| 16k | 8 | 25% | 108.761s | 87.387s | +19.65% | +4.91% |
| 4612 |
| Per-token timing unavailable (not a request failure) |
| hf-029 | — | — | — | 1859 | candidate output below tolerance |
| hf-018 | — | — | — | 2913 | Per-token timing unavailable (not a request failure) |
| hf-012 | — | — | — | 3273 | Per-token timing unavailable (not a request failure) |
| hf-010 | — | — | — | 3254 | candidate output below tolerance |
| hf-028 | — | — | — | 3355 | Per-token timing unavailable (not a request failure) |
| hf-000 | — | — | — | 2347 | candidate output below tolerance |
| hf-031 | — | — | — | 4039 | candidate output below tolerance |
| hf-016 | — | — | — | 2781 | candidate output below tolerance |
| hf-004 | — | — | — | 3463 | Per-token timing unavailable (not a request failure) |
| hf-005 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-002 | — | — | — | 3357 | Per-token timing unavailable (not a request failure) |
| hf-008 | — | — | — | 2846 | candidate output below tolerance |
| hf-020 | — | — | — | 2744 | candidate output below tolerance |
| hf-011 | — | — | — | 3842 | candidate output below tolerance |
| hf-025 | — | — | — | 2930 | Per-token timing unavailable (not a request failure) |
| hf-009 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-007 | — | — | — | 3410 | Per-token timing unavailable (not a request failure) |
| hf-024 | — | — | — | 2914 | candidate output below tolerance |
| hf-022 | — | — | — | 2703 | candidate output below tolerance |
| hf-003 | — | — | — | 3756 | candidate output below tolerance |
| hf-006 | — | — | — | 3965 | candidate output below tolerance |
| hf-021 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-023 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-026 | — | — | — | 4412 | candidate output below tolerance |
| hf-015 | — | — | — | 4920 | Per-token timing unavailable (not a request failure) |
| hf-030 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-019 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-001 | — | — | — | 4166 | Per-token timing unavailable (not a request failure) |
| 80.448s |
| 5120 |
| 39.38ms |
| 2262 gaps for 5120 tokens |
| hf-003 | 709.89ms | 65.671s | 3756 | 40.22ms | 1758 gaps for 3756 tokens |
| hf-004 | 709.01ms | 72.787s | 4868 | 39.99ms | 2184 gaps for 4868 tokens |
| hf-005 | 716.53ms | 81.534s | 5120 | 39.46ms | 2308 gaps for 5120 tokens |
| hf-006 | 713.32ms | 61.255s | 3965 | 39.35ms | 1707 gaps for 3965 tokens |
| hf-007 | 715.07ms | 61.465s | 3410 | 40.10ms | 1633 gaps for 3410 tokens |
| hf-008 | 711.49ms | 42.895s | 2846 | 40.38ms | 1118 gaps for 2846 tokens |
| hf-009 | 715.23ms | 85.614s | 5120 | 40.16ms | 2594 gaps for 5120 tokens |
| hf-010 | 715.69ms | 43.830s | 3254 | 39.40ms | 1167 gaps for 3254 tokens |
| hf-011 | 714.98ms | 68.689s | 3842 | 40.57ms | 1850 gaps for 3842 tokens |
| hf-012 | 709.41ms | 82.438s | 5120 | 40.09ms | 2349 gaps for 5120 tokens |
| hf-013 | 710.11ms | 75.103s | 5120 | 39.54ms | 2306 gaps for 5120 tokens |
| hf-014 | 713.16ms | 82.616s | 5120 | 40.18ms | 2358 gaps for 5120 tokens |
| hf-015 | 707.29ms | 83.372s | 5120 | 40.70ms | 2401 gaps for 5120 tokens |
| hf-016 | 710.80ms | 49.397s | 2781 | 40.83ms | 1295 gaps for 2781 tokens |
| hf-018 | 710.26ms | 81.634s | 5120 | 40.24ms | 2313 gaps for 5120 tokens |
| hf-019 | 713.54ms | 79.086s | 5120 | 41.61ms | 2156 gaps for 5120 tokens |
| hf-020 | 705.35ms | 58.032s | 2744 | 40.45ms | 1535 gaps for 2744 tokens |
| hf-021 | 718.98ms | 83.414s | 5120 | 40.77ms | 2295 gaps for 5120 tokens |
| hf-022 | 704.34ms | 41.158s | 2703 | 39.88ms | 1089 gaps for 2703 tokens |
| hf-023 | 702.62ms | 81.935s | 5120 | 40.41ms | 2327 gaps for 5120 tokens |
| hf-024 | 716.36ms | 45.849s | 2914 | 41.37ms | 1197 gaps for 2914 tokens |
| hf-025 | 705.17ms | 66.882s | 3878 | 39.73ms | 1929 gaps for 3878 tokens |
| hf-026 | 704.39ms | 69.845s | 4412 | 40.08ms | 1886 gaps for 4412 tokens |
| hf-027 | 704.49ms | 69.715s | 4201 | 40.55ms | 1882 gaps for 4201 tokens |
| hf-028 | 699.17ms | 76.002s | 4487 | 40.40ms | 2085 gaps for 4487 tokens |
| hf-029 | 707.70ms | 52.529s | 1859 | 40.49ms | 1381 gaps for 1859 tokens |
| hf-030 | 701.66ms | 71.083s | 5120 | 40.20ms | 2109 gaps for 5120 tokens |
| hf-031 | 701.67ms | 65.983s | 4039 | 39.59ms | 1893 gaps for 4039 tokens |
| Baseline char-16 |
|---|
| Drop |
|---|
| Limit |
|---|
| Result |
|---|
| hf-000 | 0.9496 | 0.9797 | 0.0301 | 0.1000 | Passed |
| hf-001 | 0.9591 | 0.9734 | 0.0143 | 0.1000 | Passed |
| hf-002 | 0.9645 | 0.8976 | -0.0668 | 0.1000 | Passed |
| hf-003 | 0.9926 | 0.9648 | -0.0277 | 0.1000 | Passed |
| hf-004 | 0.9424 | 0.8189 | -0.1235 | 0.1000 | Passed |
| hf-005 | 0.9538 | 0.9574 | 0.0036 | 0.1000 | Passed |
| hf-006 | 0.9863 | 0.9489 | -0.0375 | 0.1000 | Passed |
| hf-007 | 0.9872 | 0.9580 | -0.0292 | 0.1000 | Passed |
| hf-008 | 0.9535 | 0.9628 | 0.0094 | 0.1000 | Passed |
| hf-009 | 0.9842 | 0.9419 | -0.0423 | 0.1000 | Passed |
| hf-010 | 0.6456 | 0.9579 | 0.3123 | 0.1000 | Failed distinct 16-gram ratio 0.646 is more than 0.10 below baseline 0.958 over 15835 chars |
| hf-011 | 0.9697 | 0.9170 | -0.0527 | 0.1000 | Passed |
| hf-012 | 0.9804 | 0.9044 | -0.0761 | 0.1000 | Passed |
| hf-013 | 0.9737 | 0.7790 | -0.1947 | 0.1000 | Passed |
| hf-014 | 0.8873 | 0.8688 | -0.0185 | 0.1000 | Passed |
| hf-015 | 0.9447 | 0.9623 | 0.0176 | 0.1000 | Passed |
| hf-016 | 0.9584 | 0.9566 | -0.0018 | 0.1000 | Passed |
| hf-018 | 0.9403 | 0.9036 | -0.0367 | 0.1000 | Passed |
| hf-019 | 0.9103 | 0.9584 | 0.0482 | 0.1000 | Passed |
| hf-020 | 0.9918 | 0.9657 | -0.0261 | 0.1000 | Passed |
| hf-021 | 0.9373 | 0.9434 | 0.0061 | 0.1000 | Passed |
| hf-022 | 0.9192 | 0.8636 | -0.0557 | 0.1000 | Passed |
| hf-023 | 0.9741 | 0.7941 | -0.1800 | 0.1000 | Passed |
| hf-024 | 0.9272 | 0.9447 | 0.0174 | 0.1000 | Passed |
| hf-025 | 0.9866 | 0.9026 | -0.0840 | 0.1000 | Passed |
| hf-026 | 0.9491 | 0.9368 | -0.0123 | 0.1000 | Passed |
| hf-027 | 0.9753 | 0.9064 | -0.0689 | 0.1000 | Passed |
| hf-028 | 0.9634 | 0.9292 | -0.0342 | 0.1000 | Passed |
| hf-029 | 0.9662 | 0.9555 | -0.0108 | 0.1000 | Passed |
| hf-030 | 0.9489 | 0.9266 | -0.0223 | 0.1000 | Passed |
| hf-031 | 0.9600 | 0.8198 | -0.1402 | 0.1000 | Passed |
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-013 | 1 | 0.233417 | 1 | 0 |
| hf-016 | 1 | 0.983149 | 1 | 0 |
| hf-031 | 1 | 0.722809 | 1 | 0 |
| hf-030 | 1 | 0.452432 | 1 | 0 |
| hf-024 | 1 | 0.649567 | 1 | 0 |
| hf-004 | 1 | 0.27719 | 1 | 0 |
| hf-029 | 1 | 0.94331 | 1 | 0 |
| hf-006 | 1 | 0.949 | 1 | 0 |
| hf-011 | 1 | 0.750363 | 1 | 0 |
| hf-007 | 1 | 0.912541 | 1 | 0 |
| hf-027 | 1 | 0.549717 | 1 | 0 |
| hf-019 | 1 | 0.104092 | 1 | 0 |
| hf-000 | 1 | 0.907409 | 1 | 0 |
| hf-003 | 1 | 0.577771 | 1 | 0 |
| hf-022 | 1 | 0.38995 | 1 | 0 |
| hf-028 | 1 | 0.489132 | 1 | 0 |
| hf-021 | 1 | 0.488697 | 1 | 0 |
| hf-014 | 1 | 0.393375 | 1 | 0 |
| hf-012 | 1 | 0.617887 | 1 | 0 |
| hf-015 | 1 | 0.139166 | 1 | 0 |
| hf-005 | 1 | 0.382033 | 1 | 0 |
| hf-026 | 1 | 0.241993 | 1 | 0 |
| hf-010 | 1 | 0.347258 | 1 | 0 |
| hf-008 | 1 | 0.36576 | 1 | 0 |
| hf-023 | 1 | 0.405127 | 1 | 0 |
| hf-009 | 1 | 0.932255 | 1 | 0 |
| hf-002 | 1 | 0.6371 | 1 | 0 |
| hf-020 | 1 | 0.971895 | 1 | 0 |
| hf-001 | 1 | 0.829672 | 1 | 0 |
| hf-018 | 1 | 0.566662 | 1 | 0 |
| hf-025 | 1 | 0.722464 | 1 | 0 |
| hf-020 | 2 | 0.971895 | 1 | 0 |
| hf-027 | 2 | 0.549717 | 1 | 0 |
| hf-021 | 2 | 0.488697 | 1 | 0 |
| hf-008 | 2 | 0.36576 | 1 | 0 |
| hf-003 | 2 | 0.577771 | 1 | 0 |
| hf-000 | 2 | 0.907409 | 1 | 0 |
| hf-024 | 2 | 0.649567 | 1 | 0 |
| hf-026 | 2 | 0.241993 | 1 | 0 |
| hf-014 | 2 | 0.393375 | 1 | 0 |
| hf-022 | 2 | 0.38995 | 1 | 0 |
| hf-010 | 2 | 0.347258 | 1 | 0 |
| hf-007 | 2 | 0.912541 | 1 | 0 |
| hf-028 | 2 | 0.489132 | 1 | 0 |
| hf-016 | 2 | 0.983149 | 1 | 0 |
| hf-006 | 2 | 0.949 | 1 | 0 |
| hf-023 | 2 | 0.405127 | 1 | 0 |
| hf-031 | 2 | 0.722809 | 1 | 0 |
| hf-018 | 2 | 0.566662 | 1 | 0 |
| hf-025 | 2 | 0.722464 | 1 | 0 |
| hf-030 | 2 | 0.452432 | 1 | 0 |
| hf-012 | 2 | 0.617887 | 1 | 0 |
| hf-019 | 2 | 0.104092 | 1 | 0 |
| hf-004 | 2 | 0.27719 | 1 | 0 |
| hf-001 | 2 | 0.829672 | 1 | 0 |
| hf-005 | 2 | 0.382033 | 1 | 0 |
| hf-029 | 2 | 0.94331 | 1 | 0 |
| hf-013 | 2 | 0.233417 | 1 | 0 |
| hf-015 | 2 | 0.139166 | 1 | 0 |
| hf-002 | 2 | 0.6371 | 1 | 0 |
| hf-011 | 2 | 0.750363 | 1 | 0 |
| hf-009 | 2 | 0.932255 | 1 | 0 |
| hf-022 | 3 | 0.38995 | 1 | 0 |
| hf-010 | 3 | 0.347258 | 1 | 0 |
| hf-000 | 3 | 0.907409 | 1 | 0 |
| hf-008 | 3 | 0.36576 | 1 | 0 |
| hf-016 | 3 | 0.983149 | 1 | 0 |
| hf-029 | 3 | 0.94331 | 1 | 0 |
| hf-020 | 3 | 0.971895 | 1 | 0 |
| hf-007 | 3 | 0.912541 | 1 | 0 |
| hf-025 | 3 | 0.722464 | 1 | 0 |
| hf-001 | 3 | 0.829672 | 1 | 0 |
| hf-003 | 3 | 0.577771 | 1 | 0 |
| hf-011 | 3 | 0.750363 | 1 | 0 |
| hf-027 | 3 | 0.549717 | 1 | 0 |
| hf-026 | 3 | 0.241993 | 1 | 0 |
| hf-031 | 3 | 0.722809 | 1 | 0 |
| hf-028 | 3 | 0.489132 | 1 | 0 |
| hf-002 | 3 | 0.6371 | 1 | 0 |
| hf-006 | 3 | 0.949 | 1 | 0 |
| hf-013 | 3 | 0.233417 | 1 | 0 |
| hf-005 | 3 | 0.382033 | 1 | 0 |
| hf-018 | 3 | 0.566662 | 1 | 0 |
| hf-023 | 3 | 0.405127 | 1 | 0 |
| hf-012 | 3 | 0.617887 | 1 | 0 |
| hf-014 | 3 | 0.393375 | 1 | 0 |
| hf-024 | 3 | 0.649567 | 1 | 0 |
| hf-019 | 3 | 0.104092 | 1 | 0 |
| hf-015 | 3 | 0.139166 | 1 | 0 |
| hf-030 | 3 | 0.452432 | 1 | 0 |
| hf-021 | 3 | 0.488697 | 1 | 0 |
| hf-004 | 3 | 0.27719 | 1 | 0 |
| hf-009 | 3 | 0.932255 | 1 | 0 |
verdict
evidence
min logprob
num prompts
mean logprob
coverage ratio
quantile logprob
num positions scored
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
Patch hash