Round score
0.274
weighted tier speedup after failure cap and deduction
Eligible requests
29
trusted baseline exclusions removed
Failed eligible requests
6
each request counted once
Raw weighted speedup
+29.47%
before failure cap and deduction
Raw weighted speedup
Eligible speedup
Request failures cap speed credit at zero: min(0.294672, 0) = 0.294672. Negative speedups remain negative.
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 |
|---|---|---|---|---|---|---|
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.
29 prompts
| Prompt | Diagnostic speedup | Baseline | Candidate | Aligned output | Diagnostic / failure |
|---|---|---|---|---|---|
| hf-001 | — | — | — | 3618 | candidate output below tolerance |
| hf-030 | — | — | — | 3262 | Per-token timing unavailable (not a request failure) |
| hf-022 | — | — | — |
29 requests
| Request | TTFT | Total | Tokens | Slowest gap | Stream |
|---|---|---|---|---|---|
| hf-000 | 910.68ms | 45.642s | 2992 | 1.022s | 993 gaps for 2992 tokens |
| hf-001 | 2.090s | 60.781s | 3618 | 53.15ms | 1355 gaps for 3618 tokens |
| hf-002 | 1.843s | 69.086s |
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 | 29 | 29 | 25.02 |
| 2 | 0.000s | C32 | 29 | 29 | 24.86 |
| 3 | 0.000s | C32 | 29 | 29 | 24.46 |
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 |
|---|
| 16k |
|---|
| 8 |
| 25% |
| 121.650s |
| 83.250s |
| +31.57% |
| +7.89% |
| 32k | 8 | 25% | 116.561s | 84.294s | +27.68% | +6.92% |
|---|
| 64k | 7 | 25% | 121.318s | 82.933s | +31.64% | +7.91% |
|---|
| 128k | 6 | 25% | 120.339s | 87.871s | +26.98% | +6.75% |
|---|
| 4240 |
| Per-token timing unavailable (not a request failure) |
| hf-006 | — | — | — | 3326 | Per-token timing unavailable (not a request failure) |
| hf-027 | — | — | — | 3420 | Per-token timing unavailable (not a request failure) |
| hf-017 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-016 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-013 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-000 | — | — | — | 2992 | Per-token timing unavailable (not a request failure) |
| hf-021 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-020 | — | — | — | 3988 | candidate output below tolerance |
| hf-003 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-005 | — | — | — | 3472 | Per-token timing unavailable (not a request failure) |
| hf-011 | — | — | — | 3510 | candidate output below tolerance |
| hf-029 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-009 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-018 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-019 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-002 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-007 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-023 | — | — | — | 3983 | candidate output below tolerance |
| hf-028 | — | — | — | 3148 | candidate output below tolerance |
| hf-012 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-026 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-010 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-025 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-031 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-014 | — | — | — | 5120 | Per-token timing unavailable (not a request failure) |
| hf-024 | — | — | — | 3897 | candidate output below tolerance |
| 5120 |
| 84.99ms |
| 1564 gaps for 5120 tokens |
| hf-003 | 2.085s | 70.265s | 5120 | 51.59ms | 1604 gaps for 5120 tokens |
| hf-005 | 1.844s | 73.798s | 4908 | 87.47ms | 1742 gaps for 4908 tokens |
| hf-006 | 1.839s | 74.229s | 4543 | 84.94ms | 1765 gaps for 4543 tokens |
| hf-007 | 1.841s | 73.383s | 5120 | 87.04ms | 1722 gaps for 5120 tokens |
| hf-009 | 1.837s | 74.229s | 5120 | 85.58ms | 1765 gaps for 5120 tokens |
| hf-010 | 1.843s | 73.495s | 5120 | 85.93ms | 1727 gaps for 5120 tokens |
| hf-011 | 2.081s | 61.424s | 3510 | 51.16ms | 1371 gaps for 3510 tokens |
| hf-012 | 2.083s | 72.896s | 5120 | 51.50ms | 1697 gaps for 5120 tokens |
| hf-013 | 2.103s | 74.685s | 5120 | 47.47ms | 1786 gaps for 5120 tokens |
| hf-014 | 1.908s | 70.496s | 5120 | 49.31ms | 1549 gaps for 5120 tokens |
| hf-016 | 2.104s | 75.331s | 5120 | 47.48ms | 1831 gaps for 5120 tokens |
| hf-017 | 1.910s | 68.335s | 5120 | 49.48ms | 1496 gaps for 5120 tokens |
| hf-018 | 1.891s | 71.606s | 5120 | 46.54ms | 1653 gaps for 5120 tokens |
| hf-019 | 1.887s | 69.166s | 5120 | 46.85ms | 1567 gaps for 5120 tokens |
| hf-020 | 1.912s | 76.003s | 3988 | 49.33ms | 1724 gaps for 3988 tokens |
| hf-021 | 2.097s | 74.018s | 5120 | 47.99ms | 1751 gaps for 5120 tokens |
| hf-022 | 1.910s | 70.083s | 5120 | 49.49ms | 1539 gaps for 5120 tokens |
| hf-023 | 1.885s | 60.806s | 3983 | 47.15ms | 1352 gaps for 3983 tokens |
| hf-024 | 1.883s | 58.031s | 3897 | 47.25ms | 1284 gaps for 3897 tokens |
| hf-025 | 1.878s | 68.562s | 5120 | 47.20ms | 1548 gaps for 5120 tokens |
| hf-026 | 2.092s | 68.799s | 5120 | 47.96ms | 1558 gaps for 5120 tokens |
| hf-027 | 1.876s | 67.288s | 4476 | 47.13ms | 1513 gaps for 4476 tokens |
| hf-028 | 1.905s | 57.524s | 3148 | 49.49ms | 1236 gaps for 3148 tokens |
| hf-029 | 1.902s | 73.394s | 5120 | 49.97ms | 1625 gaps for 5120 tokens |
| hf-030 | 1.901s | 55.265s | 3576 | 50.13ms | 1183 gaps for 3576 tokens |
| hf-031 | 2.091s | 74.738s | 5120 | 47.94ms | 1790 gaps for 5120 tokens |
| Baseline char-16 |
|---|
| Drop |
|---|
| Limit |
|---|
| Result |
|---|
| hf-000 | 0.7571 | 0.6532 | -0.1039 | 0.1000 | Passed |
| hf-001 | 0.8735 | 0.7253 | -0.1482 | 0.1000 | Passed |
| hf-002 | 0.7297 | 0.6755 | -0.0542 | 0.1000 | Passed |
| hf-003 | 0.6889 | 0.6417 | -0.0472 | 0.1000 | Passed |
| hf-005 | 0.8350 | 0.8205 | -0.0146 | 0.1000 | Passed |
| hf-006 | 0.9106 | 0.9420 | 0.0314 | 0.1000 | Passed |
| hf-007 | 0.7489 | 0.5829 | -0.1660 | 0.1000 | Passed |
| hf-009 | 0.8735 | 0.7753 | -0.0982 | 0.1000 | Passed |
| hf-010 | 0.8271 | 0.7463 | -0.0808 | 0.1000 | Passed |
| hf-011 | 0.9321 | 0.6971 | -0.2349 | 0.1000 | Passed |
| hf-012 | 0.8210 | 0.7874 | -0.0336 | 0.1000 | Passed |
| hf-013 | 0.8969 | 0.8404 | -0.0565 | 0.1000 | Passed |
| hf-014 | 0.6950 | 0.7632 | 0.0682 | 0.1000 | Passed |
| hf-016 | 0.8087 | 0.8207 | 0.0120 | 0.1000 | Passed |
| hf-017 | 0.6697 | 0.7281 | 0.0583 | 0.1000 | Passed |
| hf-018 | 0.6722 | 0.6505 | -0.0218 | 0.1000 | Passed |
| hf-019 | 0.7714 | 0.7296 | -0.0419 | 0.1000 | Passed |
| hf-020 | 0.9616 | 0.8612 | -0.1004 | 0.1000 | Passed |
| hf-021 | 0.8267 | 0.7980 | -0.0287 | 0.1000 | Passed |
| hf-022 | 0.6442 | 0.8596 | 0.2154 | 0.1000 | Failed distinct 16-gram ratio 0.644 is more than 0.10 below baseline 0.860 over 21079 chars |
| hf-023 | 0.7318 | 0.7329 | 0.0011 | 0.1000 | Passed |
| hf-024 | 0.7087 | 0.7422 | 0.0336 | 0.1000 | Passed |
| hf-025 | 0.6925 | 0.7315 | 0.0390 | 0.1000 | Passed |
| hf-026 | 0.7512 | 0.8155 | 0.0643 | 0.1000 | Passed |
| hf-027 | 0.7954 | 0.5957 | -0.1998 | 0.1000 | Passed |
| hf-028 | 0.9011 | 0.7612 | -0.1398 | 0.1000 | Passed |
| hf-029 | 0.7626 | 0.7455 | -0.0171 | 0.1000 | Passed |
| hf-030 | 0.7713 | 0.8648 | 0.0934 | 0.1000 | Passed |
| hf-031 | 0.9054 | 0.8597 | -0.0457 | 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-005 | 1 | 0.760143 | 1 | 0 |
| hf-024 | 1 | 0.121643 | 1 | 0 |
| hf-000 | 1 | 0.101237 | 1 | 0 |
| hf-018 | 1 | 0.221058 | 1 | 0 |
| hf-006 | 1 | 0.8152 | 1 | 0 |
| hf-030 | 1 | 0.476485 | 1 | 0 |
| hf-001 | 1 | 0.444844 | 1 | 0 |
| hf-011 | 1 | 0.548234 | 1 | 0 |
| hf-028 | 1 | 0.165048 | 1 | 0 |
| hf-002 | 1 | 0.426332 | 1 | 0 |
| hf-019 | 1 | 0.382532 | 1 | 0 |
| hf-025 | 1 | 0.298822 | 1 | 0 |
| hf-026 | 1 | 0.240206 | 1 | 0 |
| hf-010 | 1 | 0.319315 | 1 | 0 |
| hf-003 | 1 | 0.740817 | 1 | 0 |
| hf-029 | 1 | 0.316805 | 1 | 0 |
| hf-027 | 1 | 0.901469 | 1 | 0 |
| hf-022 | 1 | 0.985248 | 1 | 0 |
| hf-007 | 1 | 0.412296 | 1 | 0 |
| hf-012 | 1 | 0.350242 | 1 | 0 |
| hf-023 | 1 | 0.485263 | 1 | 0 |
| hf-014 | 1 | 0.248189 | 1 | 0 |
| hf-017 | 1 | 0.913776 | 1 | 0 |
| hf-021 | 1 | 0.504238 | 1 | 0 |
| hf-013 | 1 | 0.726073 | 1 | 0 |
| hf-031 | 1 | 0.106656 | 1 | 0 |
| hf-009 | 1 | 0.602937 | 1 | 0 |
| hf-016 | 1 | 0.940146 | 1 | 0 |
| hf-020 | 1 | 0.921463 | 1 | 0 |
| hf-023 | 2 | 0.485263 | 1 | 0 |
| hf-000 | 2 | 0.101237 | 1 | 0 |
| hf-027 | 2 | 0.901469 | 1 | 0 |
| hf-030 | 2 | 0.476485 | 1 | 0 |
| hf-028 | 2 | 0.165048 | 1 | 0 |
| hf-001 | 2 | 0.444844 | 1 | 0 |
| hf-017 | 2 | 0.913776 | 1 | 0 |
| hf-002 | 2 | 0.426332 | 1 | 0 |
| hf-022 | 2 | 0.985248 | 1 | 0 |
| hf-014 | 2 | 0.248189 | 1 | 0 |
| hf-011 | 2 | 0.548234 | 1 | 0 |
| hf-003 | 2 | 0.740817 | 1 | 0 |
| hf-019 | 2 | 0.382532 | 1 | 0 |
| hf-029 | 2 | 0.316805 | 1 | 0 |
| hf-021 | 2 | 0.504238 | 1 | 0 |
| hf-025 | 2 | 0.298822 | 1 | 0 |
| hf-009 | 2 | 0.602937 | 1 | 0 |
| hf-018 | 2 | 0.221058 | 1 | 0 |
| hf-026 | 2 | 0.240206 | 1 | 0 |
| hf-012 | 2 | 0.350242 | 1 | 0 |
| hf-020 | 2 | 0.921463 | 1 | 0 |
| hf-013 | 2 | 0.726073 | 1 | 0 |
| hf-016 | 2 | 0.940146 | 1 | 0 |
| hf-007 | 2 | 0.412296 | 1 | 0 |
| hf-010 | 2 | 0.319315 | 1 | 0 |
| hf-031 | 2 | 0.106656 | 1 | 0 |
| hf-024 | 2 | 0.121643 | 1 | 0 |
| hf-006 | 2 | 0.8152 | 1 | 0 |
| hf-005 | 2 | 0.760143 | 1 | 0 |
| hf-022 | 3 | 0.985248 | 1 | 0 |
| hf-000 | 3 | 0.101237 | 1 | 0 |
| hf-017 | 3 | 0.913776 | 1 | 0 |
| hf-028 | 3 | 0.165048 | 1 | 0 |
| hf-030 | 3 | 0.476485 | 1 | 0 |
| hf-011 | 3 | 0.548234 | 1 | 0 |
| hf-024 | 3 | 0.121643 | 1 | 0 |
| hf-023 | 3 | 0.485263 | 1 | 0 |
| hf-026 | 3 | 0.240206 | 1 | 0 |
| hf-027 | 3 | 0.901469 | 1 | 0 |
| hf-003 | 3 | 0.740817 | 1 | 0 |
| hf-012 | 3 | 0.350242 | 1 | 0 |
| hf-014 | 3 | 0.248189 | 1 | 0 |
| hf-025 | 3 | 0.298822 | 1 | 0 |
| hf-002 | 3 | 0.426332 | 1 | 0 |
| hf-019 | 3 | 0.382532 | 1 | 0 |
| hf-001 | 3 | 0.444844 | 1 | 0 |
| hf-018 | 3 | 0.221058 | 1 | 0 |
| hf-016 | 3 | 0.940146 | 1 | 0 |
| hf-031 | 3 | 0.106656 | 1 | 0 |
| hf-013 | 3 | 0.726073 | 1 | 0 |
| hf-007 | 3 | 0.412296 | 1 | 0 |
| hf-010 | 3 | 0.319315 | 1 | 0 |
| hf-005 | 3 | 0.760143 | 1 | 0 |
| hf-006 | 3 | 0.8152 | 1 | 0 |
| hf-009 | 3 | 0.602937 | 1 | 0 |
| hf-029 | 3 | 0.316805 | 1 | 0 |
| hf-021 | 3 | 0.504238 | 1 | 0 |
| hf-020 | 3 | 0.921463 | 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