Round 2 breakdown

disqualifiedchallengerSubmission

Round score

—

degenerate output (baseline-relative repetition failures exceed allowance of 4 prompts (hf-028: distinct 16-gram ratio 0.432 is more than 0.10 below baseline 0.761 over 43859 chars))

Eligible requests

—

trusted baseline exclusions removed

Failed eligible requests

—

each request counted once

Raw weighted speedup

—

before failure cap and deduction

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-028: distinct 16-gram ratio 0.432 is more than 0.10 below baseline 0.761 over 43859 chars))

Request trace

29 requests

Request	TTFT	Total	Tokens	Slowest gap	Stream
hf-000	1.212s	78.365s	4001	15.180s	3104 gaps for 4001 tokens
hf-001	5.603s	56.822s	3816	10.985s	2109 gaps for 3816 tokens
hf-002	6.337s	70.829s

Observed request concurrency

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	16.77
2	0.000s	C32	29	29	16.80
3	0.000s	C32	29	29	16.71

Per-prompt repetition checks

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

hf-000	0.9512	0.6532	-0.2980	0.1000	Passed
hf-001	0.9421	0.7253	-0.2168	0.1000	Passed
hf-002	0.7768	0.6755	-0.1013	0.1000	Passed
hf-003	0.8994	0.6417	-0.2577	0.1000	Passed
hf-005	0.8029	0.8205	0.0176	0.1000	Passed
hf-006	0.8376	0.9420	0.1044	0.1000	Failed distinct 16-gram ratio 0.838 is more than 0.10 below baseline 0.942 over 22713 chars
hf-007	0.9452	0.5829	-0.3623	0.1000	Passed
hf-009	0.8579	0.7753	-0.0825	0.1000	Passed
hf-010	0.8954	0.7463	-0.1491	0.1000	Passed
hf-011	0.6356	0.6971	0.0615	0.1000	Passed
hf-012	0.9714	0.7874	-0.1840	0.1000	Passed
hf-013	0.7724	0.8404	0.0680	0.1000	Passed
hf-014	0.9063	0.7632	-0.1431	0.1000	Passed
hf-016	0.4055	0.8207	0.4152	0.1000	Failed distinct 16-gram ratio 0.406 is more than 0.10 below baseline 0.821 over 22799 chars
hf-017	0.7422	0.7281	-0.0141	0.1000	Passed
hf-018	0.8774	0.6505	-0.2270	0.1000	Passed
hf-019	0.9251	0.7296	-0.1956	0.1000	Passed
hf-020	0.9708	0.8612	-0.1096	0.1000	Passed
hf-021	0.4699	0.7980	0.3282	0.1000	Failed distinct 16-gram ratio 0.470 is more than 0.10 below baseline 0.798 over 21070 chars
hf-022	0.7541	0.8596	0.1055	0.1000	Failed distinct 16-gram ratio 0.754 is more than 0.10 below baseline 0.860 over 20393 chars
hf-023	0.9264	0.7329	-0.1935	0.1000	Passed
hf-024	0.9088	0.7422	-0.1666	0.1000	Passed
hf-025	0.9119	0.7315	-0.1804	0.1000	Passed
hf-026	0.9565	0.8155	-0.1410	0.1000	Passed
hf-027	0.9060	0.5957	-0.3104	0.1000	Passed
hf-028	0.4321	0.7612	0.3291	0.1000	Failed distinct 16-gram ratio 0.432 is more than 0.10 below baseline 0.761 over 43859 chars
hf-029	0.9672	0.7455	-0.2217	0.1000	Passed
hf-030	0.3346	0.8648	0.5301	0.1000	Failed distinct 16-gram ratio 0.335 is more than 0.10 below baseline 0.865 over 16995 chars
hf-031	0.9539	0.8597	-0.0942	0.1000	Passed

Prompt	Repetition	Temperature	Top p
hf-001	1	0.444844	1
hf-003	1	0.740817	1
hf-006	1	0.8152	1
hf-005	1	0.760143	1
hf-002	1	0.426332	1
hf-007	1	0.412296	1
hf-000	1	0.101237	1
hf-010	1	0.319315	1
hf-014	1	0.248189	1
hf-009	1	0.602937	1
hf-011	1	0.548234	1
hf-013	1	0.726073	1
hf-012	1	0.350242	1
hf-016	1	0.940146	1
hf-017	1	0.913776	1
hf-022	1	0.985248	1
hf-020	1	0.921463	1
hf-018	1	0.221058	1
hf-024	1	0.121643	1
hf-019	1	0.382532	1
hf-023	1	0.485263	1
hf-021	1	0.504238	1
hf-028	1	0.165048	1
hf-030	1	0.476485	1
hf-027	1	0.901469	1
hf-026	1	0.240206	1
hf-029	1	0.316805	1
hf-025	1	0.298822	1
hf-031	1	0.106656	1
hf-005	2	0.760143	1
hf-006	2	0.8152	1
hf-001	2	0.444844	1
hf-003	2	0.740817	1
hf-007	2	0.412296	1
hf-002	2	0.426332	1
hf-010	2	0.319315	1
hf-009	2	0.602937	1
hf-000	2	0.101237	1
hf-011	2	0.548234	1
hf-014	2	0.248189	1
hf-012	2	0.350242	1
hf-013	2	0.726073	1
hf-016	2	0.940146	1
hf-022	2	0.985248	1
hf-020	2	0.921463	1
hf-017	2	0.913776	1
hf-019	2	0.382532	1
hf-018	2	0.221058	1
hf-028	2	0.165048	1
hf-021	2	0.504238	1
hf-030	2	0.476485	1
hf-023	2	0.485263	1
hf-024	2	0.121643	1
hf-027	2	0.901469	1
hf-026	2	0.240206	1
hf-025	2	0.298822	1
hf-031	2	0.106656	1
hf-029	2	0.316805	1
hf-003	3	0.740817	1
hf-006	3	0.8152	1
hf-005	3	0.760143	1
hf-002	3	0.426332	1
hf-001	3	0.444844	1
hf-007	3	0.412296	1
hf-000	3	0.101237	1
hf-010	3	0.319315	1
hf-009	3	0.602937	1
hf-018	3	0.221058	1
hf-013	3	0.726073	1
hf-014	3	0.248189	1
hf-011	3	0.548234	1
hf-012	3	0.350242	1
hf-022	3	0.985248	1
hf-016	3	0.940146	1
hf-017	3	0.913776	1
hf-020	3	0.921463	1
hf-019	3	0.382532	1
hf-023	3	0.485263	1
hf-021	3	0.504238	1
hf-030	3	0.476485	1
hf-028	3	0.165048	1
hf-027	3	0.901469	1
hf-024	3	0.121643	1
hf-026	3	0.240206	1
hf-029	3	0.316805	1
hf-025	3	0.298822	1
hf-031	3	0.106656	1

Round 2 breakdown

Request trace

Observed request concurrency

Per-prompt repetition checks

Round 2 breakdown

Round 2 breakdown

Request trace

Observed request concurrency

Per-prompt repetition checks

Replay sampling

Correctness

SLA replay

Engine