Duel

Crown held5 Nov 2026, 01:41 UTC3b004616da…852278

Reigning model

Holds the crown

domFree/attune-j4

Revision: 4a2651b08cb70332bfe9b86ef9b2efaff1825eb8
Submission key: 12d59031ac1dcdc3

85.4%

Average over the tasks

Beat block hammer
Move stapler pad
Place bread skillet
Place burger fries
Place can basket
Place cans plasticbox
Place container plate
Place dual shoes
Place empty cup
Place fan
Place mouse pad
Place object basket
Place object scale
Place object stand
Place phone stand
Place shoe

Challenger

maginion/king-3

Revision: bef1bd6f7626fee2610e6d73c2deaaec3043c114
Submission key: 9aadc2aae1b8a527

83.3%

Average over the tasks

Beat block hammer
Move stapler pad
Place bread skillet
Place burger fries
Place can basket
Place cans plasticbox
Place container plate
Place dual shoes
Place empty cup
Place fan
Place mouse pad
Place object basket
Place object scale
Place object stand
Place phone stand
Place shoe

Margin check83.3% − 85.4% = -2.1 ptsneeds ≥ +3.0 ptsNot met

Verdict derivation

recomputed from 240 unitsCrown held

The ledger

Per-task success rates for both sides and their difference
Task	Reigning	Challenger	Δ
Beat block hammer	60.0%9 / 15	46.7%7 / 15	-13.3 pts
Move stapler pad	73.3%11 / 15	73.3%11 / 15	0.0 pts
Place bread skillet	93.3%14 / 15	93.3%14 / 15	0.0 pts
Place burger fries	86.7%13 / 15	93.3%14 / 15	+6.7 pts
Place can basket	93.3%14 / 15	80.0%12 / 15	-13.3 pts
Place cans plasticbox	100.0%15 / 15	100.0%15 / 15	0.0 pts
Place container plate	93.3%14 / 15	100.0%15 / 15	+6.7 pts
Place dual shoes	66.7%10 / 15	53.3%8 / 15	-13.3 pts
Place empty cup	100.0%15 / 15	100.0%15 / 15	0.0 pts
Place fan	80.0%12 / 15	80.0%12 / 15	0.0 pts
Place mouse pad	66.7%10 / 15	73.3%11 / 15	+6.7 pts
Place object basket	93.3%14 / 15	86.7%13 / 15	-6.7 pts
Place object scale	93.3%14 / 15	93.3%14 / 15	0.0 pts
Place object stand	100.0%15 / 15	93.3%14 / 15	-6.7 pts
Place phone stand	80.0%12 / 15	93.3%14 / 15	+13.3 pts
Place shoe	86.7%13 / 15	73.3%11 / 15	-13.3 pts
Averageneeds ≥ 88.4%	85.4%	83.3%	-2.1 pts

challenger average ≥ reigning average + +3.0 pts

83.3% ≥ 85.4% + +3.0 pts · needs ≥ 88.4%

Not met

The reigning model kept the crown

83.3% vs 85.4% · -2.1 pts

Paired outcomes

which side solved each unit

Won

13

Lost

18

Tied

209

Decided

31

Void

0

The paired record

240 units · 31 decided · 209 tied

Reigning
60.0%9/15
Challenger
46.7%7/15
Tied
scene 403445962
Reigning
Fail
Challenger
Fail
Reigning
scene 725491657
Reigning
Pass
Challenger
Fail
Tied
scene 984713971
Reigning
Pass
Challenger
Pass
Tied
scene 163477238
Reigning
Pass
Challenger
Pass
Tied
scene 800621023
Reigning
Pass
Challenger
Pass
Reigning
scene 1476302060
Reigning
Pass
Challenger
Fail
Reigning
scene 1280851561
Reigning
Pass
Challenger
Fail
Tied
scene 438514818
Reigning
Fail
Challenger
Fail
Challenger
scene 99933709
Reigning
Fail
Challenger
Pass
Reigning
scene 1926039860
Reigning
Pass
Challenger
Fail
Tied
scene 1895440833
Reigning
Pass
Challenger
Pass
Challenger
scene 796728922
Reigning
Fail
Challenger
Pass
Tied
scene 987839730
Reigning
Fail
Challenger
Fail
Tied
scene 1462587135
Reigning
Pass
Challenger
Pass
Tied
scene 1845154318
Reigning
Fail
Challenger
Fail
Reigning
73.3%11/15
Challenger
73.3%11/15
Reigning
93.3%14/15
Challenger
93.3%14/15
Reigning
86.7%13/15
Challenger
93.3%14/15
Reigning
93.3%14/15
Challenger
80.0%12/15
Reigning
100.0%15/15
Challenger
100.0%15/15
Reigning
93.3%14/15
Challenger
100.0%15/15
Reigning
66.7%10/15
Challenger
53.3%8/15
Reigning
100.0%15/15
Challenger
100.0%15/15
Reigning
80.0%12/15
Challenger
80.0%12/15
Reigning
66.7%10/15
Challenger
73.3%11/15
Reigning
93.3%14/15
Challenger
86.7%13/15
Reigning
93.3%14/15
Challenger
93.3%14/15
Reigning
100.0%15/15
Challenger
93.3%14/15
Reigning
80.0%12/15
Challenger
93.3%14/15
Reigning
86.7%13/15
Challenger
73.3%11/15

Unit	Scene	Reigning	Challenger	Outcome
		60.0%9/15	46.7%7/15	-13.3 pts
	scene 403445962	Fail	Fail	Tied
	scene 725491657	Pass	Fail	Reigning
	scene 984713971	Pass	Pass	Tied
	scene 163477238	Pass	Pass	Tied
	scene 800621023	Pass	Pass	Tied
	scene 1476302060	Pass	Fail	Reigning
	scene 1280851561	Pass	Fail	Reigning
	scene 438514818	Fail	Fail	Tied
	scene 99933709	Fail	Pass	Challenger
	scene 1926039860	Pass	Fail	Reigning
	scene 1895440833	Pass	Pass	Tied
	scene 796728922	Fail	Pass	Challenger
	scene 987839730	Fail	Fail	Tied
	scene 1462587135	Pass	Pass	Tied
	scene 1845154318	Fail	Fail	Tied
		73.3%11/15	73.3%11/15	0.0 pts
		93.3%14/15	93.3%14/15	0.0 pts
		86.7%13/15	93.3%14/15	+6.7 pts
		93.3%14/15	80.0%12/15	-13.3 pts
		100.0%15/15	100.0%15/15	0.0 pts
		93.3%14/15	100.0%15/15	+6.7 pts
		66.7%10/15	53.3%8/15	-13.3 pts
		100.0%15/15	100.0%15/15	0.0 pts
		80.0%12/15	80.0%12/15	0.0 pts
		66.7%10/15	73.3%11/15	+6.7 pts
		93.3%14/15	86.7%13/15	-6.7 pts
		93.3%14/15	93.3%14/15	0.0 pts
		100.0%15/15	93.3%14/15	-6.7 pts
		80.0%12/15	93.3%14/15	+13.3 pts
		86.7%13/15	73.3%11/15	-13.3 pts

Provenance

Record: 3b004616daa8…af852278
Kind: Duel
Finished: 5 Nov 2026, 01:41 UTC
Wall time: 1h 47m
Units per side: 15 × 16 tasks = 240
Crown rule: challenger average ≥ reigning average + +3.0 pts
Void units: 0.0%
Duel id: 1a8a5d6afdcf…78a08ff4
Seed block: 9,436,064
Seed block hash: 0x6ac17ba1d0…c46b7dd3
Spec version: 3
Benchmark commit: e87619a41914…0a166ccc
Benchmark source: 2358200e9bdd…57aa5a09
Reigning weights: 06092389136a…2ab56c1c
Challenger weights: 4671c9b3bd45…9e52f9cb
Block: 272