Round score
0.7717
+77.17%
Prompts scored
26 / 32
median_e2e_speedup
Below tolerance
6
answered too little to score
Zeroed total
6
scored 0.0 without measuring
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.
32 prompts
| Prompt | Speedup | Baseline | Candidate | Tokens | Gate |
|---|---|---|---|---|---|
| hf-000 | +75.86% | 0.712s | 0.172s | 42 | — |
| hf-001 | +80.66% | 0.760s | 0.147s | 45 | — |
| hf-003 | +81.99% | 0.755s | 0.136s |
32 requests
| Request | TTFT | Total | Tokens | Slowest gap | Stream |
|---|---|---|---|---|---|
| hf-000 | 49.36ms | 0.172s | 42 | 15.77ms | — |
| hf-001 | 53.45ms | 0.147s | 45 | 16.31ms | — |
| hf-002 | 62.85ms | 0.191s |
verdict
evidence
min logprob
num prompts
mean logprob
coverage ratio
quantile logprob
num positions scored
role
| 44 |
| — |
| hf-002 | — | — | — | 51 | candidate output below tolerance |
| hf-004 | +81.27% | 0.730s | 0.137s | 42 | — |
| hf-005 | +78.44% | 0.753s | 0.162s | 47 | — |
| hf-006 | +78.36% | 0.694s | 0.150s | 42 | — |
| hf-007 | — | — | — | 28 | candidate output below tolerance |
| hf-009 | — | — | — | 28 | candidate output below tolerance |
| hf-008 | +72.65% | 0.942s | 0.258s | 63 | — |
| hf-010 | — | — | — | 16 | candidate output below tolerance |
| hf-012 | +73.52% | 0.580s | 0.154s | 37 | — |
| hf-011 | +78.56% | 0.903s | 0.194s | 52 | — |
| hf-013 | +68.68% | 0.497s | 0.156s | 28 | — |
| hf-014 | +82.13% | 0.739s | 0.132s | 42 | — |
| hf-015 | +78.18% | 0.717s | 0.157s | 41 | — |
| hf-016 | +81.80% | 0.739s | 0.134s | 43 | — |
| hf-017 | +76.31% | 0.742s | 0.176s | 43 | — |
| hf-018 | +79.14% | 0.704s | 0.147s | 42 | — |
| hf-019 | +79.54% | 0.872s | 0.178s | 54 | — |
| hf-020 | +79.57% | 0.662s | 0.135s | 41 | — |
| hf-021 | — | — | — | 42 | candidate output below tolerance |
| hf-023 | — | — | — | 26 | candidate output below tolerance |
| hf-022 | +79.40% | 0.653s | 0.134s | 41 | — |
| hf-024 | +78.86% | 0.701s | 0.148s | 46 | — |
| hf-026 | +69.26% | 0.359s | 0.110s | 23 | — |
| hf-025 | +77.13% | 0.591s | 0.135s | 42 | — |
| hf-029 | +64.50% | 0.325s | 0.115s | 20 | — |
| hf-028 | +74.46% | 0.708s | 0.181s | 41 | — |
| hf-027 | +78.22% | 0.812s | 0.177s | 54 | — |
| hf-030 | +77.22% | 0.586s | 0.134s | 41 | — |
| hf-031 | +71.63% | 0.297s | 0.084s | 20 | — |
| 51 |
| 18.34ms |
| — |
| hf-003 | 29.66ms | 0.136s | 44 | 15.74ms | — |
| hf-004 | 44.40ms | 0.137s | 42 | 15.77ms | — |
| hf-005 | 53.78ms | 0.162s | 48 | 15.96ms | — |
| hf-006 | 28.22ms | 0.150s | 42 | 15.58ms | — |
| hf-007 | 37.67ms | 0.130s | 28 | 15.63ms | — |
| hf-008 | 63.97ms | 0.290s | 75 | 45.02ms | — |
| hf-009 | 57.90ms | 0.137s | 28 | 16.33ms | — |
| hf-010 | 62.84ms | 0.095s | 16 | 16.29ms | — |
| hf-011 | 53.89ms | 0.194s | 52 | 16.08ms | — |
| hf-012 | 32.28ms | 0.260s | 57 | 44.49ms | — |
| hf-013 | 92.35ms | 0.156s | 28 | 16.26ms | — |
| hf-014 | 39.92ms | 0.132s | 47 | 15.51ms | — |
| hf-015 | 63.10ms | 0.157s | 41 | 16.47ms | — |
| hf-016 | 27.85ms | 0.134s | 43 | 15.89ms | — |
| hf-017 | 52.29ms | 0.176s | 43 | 15.90ms | — |
| hf-018 | 53.92ms | 0.162s | 45 | 15.97ms | — |
| hf-019 | 54.70ms | 0.178s | 54 | 15.67ms | — |
| hf-020 | 28.66ms | 0.135s | 41 | 15.61ms | — |
| hf-021 | 63.34ms | 0.175s | 42 | 18.82ms | — |
| hf-022 | 27.71ms | 0.134s | 41 | 16.41ms | — |
| hf-023 | 64.06ms | 0.130s | 26 | 17.05ms | — |
| hf-024 | 40.78ms | 0.148s | 46 | 15.45ms | — |
| hf-025 | 29.33ms | 0.135s | 42 | 15.43ms | — |
| hf-026 | 62.33ms | 0.189s | 52 | 18.20ms | — |
| hf-027 | 64.24ms | 0.192s | 63 | 18.22ms | — |
| hf-028 | 69.17ms | 0.181s | 41 | 18.65ms | — |
| hf-029 | 64.20ms | 0.177s | 45 | 18.32ms | — |
| hf-030 | 55.29ms | 0.149s | 46 | 15.97ms | — |
| hf-031 | 38.19ms | 0.131s | 43 | 15.66ms | — |
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
Patch hash