Round score
—
degenerate output (baseline-relative repetition failures exceed allowance of 4 prompts (hf-010: distinct 16-gram ratio 0.618 is more than 0.10 below baseline 0.971 over 18170 chars))
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.
No prompt scores
degenerate output (baseline-relative repetition failures exceed allowance of 4 prompts (hf-010: distinct 16-gram ratio 0.618 is more than 0.10 below baseline 0.971 over 18170 chars))
14 requests
| Request | TTFT | Total | Tokens | Slowest gap | Stream |
|---|---|---|---|---|---|
| hf-000 | 98.27ms | 15.596s | 3436 | 41.66ms | 1112 gaps for 3436 tokens |
| hf-001 | 95.77ms | 13.117s | 3072 | 15.70ms | 937 gaps for 3072 tokens |
| hf-002 | 101.74ms | 23.185s |
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 | C4 | 4 | 4 | 3.47 |
| 1 | 38.492s | C4 | 4 | 4 | 3.19 |
| 2 | 0.000s | C4 | 4 | 4 | 3.45 |
| 2 | 39.347s | C4 | 4 | 4 | 3.29 |
| 3 | 0.000s | C4 | 4 | 4 | 3.06 |
| 3 | 36.639s | C4 | 4 | 4 | 3.06 |
| 5120 |
| 76.26ms |
| 1584 gaps for 5120 tokens |
| hf-003 | 98.42ms | 19.710s | 4066 | 41.54ms | 1403 gaps for 4066 tokens |
| hf-004 | 142.70ms | 24.573s | 5120 | 42.42ms | 1684 gaps for 5120 tokens |
| hf-005 | 98.75ms | 22.272s | 4730 | 40.83ms | 1584 gaps for 4730 tokens |
| hf-007 | 70.25ms | 22.887s | 4755 | 40.86ms | 1658 gaps for 4755 tokens |
| hf-008 | 69.86ms | 21.415s | 5120 | 40.69ms | 1564 gaps for 5120 tokens |
| hf-009 | 138.99ms | 15.466s | 3517 | 16.36ms | 1061 gaps for 3517 tokens |
| hf-010 | 70.45ms | 18.566s | 4607 | 16.60ms | 1358 gaps for 4607 tokens |
| hf-011 | 139.93ms | 19.116s | 3815 | 16.38ms | 1312 gaps for 3815 tokens |
| hf-013 | 87.92ms | 18.962s | 4116 | 41.80ms | 1336 gaps for 4116 tokens |
| hf-014 | 68.92ms | 22.321s | 4808 | 41.95ms | 1627 gaps for 4808 tokens |
| hf-015 | 86.39ms | 20.309s | 3934 | 16.36ms | 1453 gaps for 3934 tokens |
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 | Baseline char-16 | Drop | Limit | Result |
|---|---|---|---|---|---|
| hf-000 | 0.9132 | 0.9800 | 0.0668 | 0.1000 | Passed |
| hf-001 | 0.8104 | 0.9523 | 0.1419 | 0.1000 | Failed distinct 16-gram ratio 0.810 is more than 0.10 below baseline 0.952 over 12639 chars |
| hf-002 | 0.8491 | 0.8879 | 0.0388 | 0.1000 | Passed |
| hf-003 | 0.8779 | 0.8584 | -0.0195 | 0.1000 | Passed |
| hf-004 | 0.8329 | 0.9443 | 0.1115 | 0.1000 | Failed distinct 16-gram ratio 0.833 is more than 0.10 below baseline 0.944 over 20621 chars |
| hf-005 | 0.8806 | 0.9511 | 0.0705 | 0.1000 | Passed |
| hf-007 | 0.7934 | 0.8173 | 0.0239 | 0.1000 | Passed |
| hf-008 | 0.6386 | 0.9108 | 0.2722 | 0.1000 | Failed longest repeated span ratio 0.320 at or above harness ceiling over 22203 chars |
| hf-009 | 0.7165 | 0.9288 | 0.2123 | 0.1000 | Failed distinct 16-gram ratio 0.716 is more than 0.10 below baseline 0.929 over 14759 chars |
| hf-010 | 0.6185 | 0.9711 | 0.3526 | 0.1000 | Failed longest repeated span ratio 0.311 at or above harness ceiling over 18170 chars |
| hf-011 | 0.9367 | 0.9402 | 0.0035 | 0.1000 | Passed |
| hf-013 | 0.9513 | 0.9677 | 0.0164 | 0.1000 | Passed |
| hf-014 | 0.9088 | 0.9505 | 0.0417 | 0.1000 | Passed |
| hf-015 | 0.9312 | 0.8941 | -0.0371 | 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-001 | 1 | 0.901395 | 1 | 0 |
| hf-000 | 1 | 0.901755 | 1 | 0 |
| hf-003 | 1 | 0.305979 | 1 | 0 |
| hf-005 | 1 | 0.615361 | 1 | 0 |
| hf-007 | 1 | 0.209852 | 1 | 0 |
| hf-010 | 1 | 0.922572 | 1 | 0 |
| hf-008 | 1 | 0.685191 | 1 | 0 |
| hf-009 | 1 | 0.915819 | 1 | 0 |
| hf-011 | 1 | 0.595321 | 1 | 0 |
| hf-002 | 1 | 0.849391 | 1 | 0 |
| hf-004 | 1 | 0.425489 | 1 | 0 |
| hf-013 | 1 | 0.96432 | 1 | 0 |
| hf-015 | 1 | 0.11923 | 1 | 0 |
| hf-014 | 1 | 0.545799 | 1 | 0 |
| hf-001 | 2 | 0.901395 | 1 | 0 |
| hf-000 | 2 | 0.901755 | 1 | 0 |
| hf-003 | 2 | 0.305979 | 1 | 0 |
| hf-005 | 2 | 0.615361 | 1 | 0 |
| hf-007 | 2 | 0.209852 | 1 | 0 |
| hf-008 | 2 | 0.685191 | 1 | 0 |
| hf-010 | 2 | 0.922572 | 1 | 0 |
| hf-011 | 2 | 0.595321 | 1 | 0 |
| hf-009 | 2 | 0.915819 | 1 | 0 |
| hf-002 | 2 | 0.849391 | 1 | 0 |
| hf-004 | 2 | 0.425489 | 1 | 0 |
| hf-013 | 2 | 0.96432 | 1 | 0 |
| hf-014 | 2 | 0.545799 | 1 | 0 |
| hf-015 | 2 | 0.11923 | 1 | 0 |
| hf-000 | 3 | 0.901755 | 1 | 0 |
| hf-001 | 3 | 0.901395 | 1 | 0 |
| hf-003 | 3 | 0.305979 | 1 | 0 |
| hf-005 | 3 | 0.615361 | 1 | 0 |
| hf-008 | 3 | 0.685191 | 1 | 0 |
| hf-007 | 3 | 0.209852 | 1 | 0 |
| hf-010 | 3 | 0.922572 | 1 | 0 |
| hf-009 | 3 | 0.915819 | 1 | 0 |
| hf-011 | 3 | 0.595321 | 1 | 0 |
| hf-013 | 3 | 0.96432 | 1 | 0 |
| hf-002 | 3 | 0.849391 | 1 | 0 |
| hf-004 | 3 | 0.425489 | 1 | 0 |
| hf-014 | 3 | 0.545799 | 1 | 0 |
| hf-015 | 3 | 0.11923 | 1 | 0 |
reason
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