Arena Night Sky Transformation
This challenge has been archived.
Results are preserved and may still appear across the site, but it's no longer actively competed.
16 models were given the same image and edit instruction, and the community voted blind on which outputs looked best.
How it works
#1 — Reve Image 1.0
Source Image
Night Sky Transformation
Edit instruction
“ Change the scene to night: a deep, dark sky with subtle, glistening stars visible behind the mountain. ”
Challenge Rankings
16 models
#
Model
Price
¢/img
Speed
Elo
1
$$
4¢
1268
2
$$$
6.7¢
1249
3
$
0.9¢
1247
4
$$
1.5¢
1241
5
$$
3.5¢
1218
6
$$
3¢
1210
7
$$$
7¢
1209
8
$$
3¢
1209
9
$$
2¢
1205
10
$$
3¢
1202
11
$$
2.2¢
1197
12
$$
3.9¢
1174
13
$$
3¢
1160
14
$$
4¢
1152
15
$$$
6¢
1145
16
$
0.5¢
1034
Reve Image 1.0 leads the leaderboard with a 1270 Elo, maintaining a 22-point gap over GPT Image 1.5, which offers a higher 58.2% win rate at a 77% lower price point. Despite its premium price tier, Nano Banana Pro trails the leader by 23 Elo points, while the budget-tier GPT Image 1.5 effectively matches its performance.
5 models waiting for enough speed data
Competitors
16 models, ranked by Elo
Reve Image 1.0
3 attempts · best result used for ranking
1268
Elo
78%W
·
11%L
·
11%T
1143
Elo
45%W
·
45%L
·
10%T
1035
Elo
17%W
·
83%L
·
0%T
Nano Banana Pro
3 attempts · best result used for ranking
1249
Elo
83%W
·
17%L
·
0%T
1213
Elo
69%W
·
13%L
·
18%T
1134
Elo
57%W
·
29%L
·
14%T
GPT Image 1.5
3 attempts · best result used for ranking
1247
Elo
54%W
·
31%L
·
15%T
1164
Elo
50%W
·
33%L
·
17%T
1157
Elo
45%W
·
45%L
·
10%T
FLUX.2 [pro]
3 attempts · best result used for ranking
1241
Elo
69%W
·
25%L
·
6%T
1076
Elo
11%W
·
67%L
·
22%T
1073
Elo
11%W
·
67%L
·
22%T
Seedream 5.0 Lite
3 attempts · best result used for ranking
1218
Elo
83%W
·
17%L
·
0%T
1210
Elo
62%W
·
23%L
·
15%T
1181
Elo
67%W
·
33%L
·
0%T
FLUX.2 [max]
3 attempts · best result used for ranking
1210
Elo
62%W
·
31%L
·
7%T
1119
Elo
50%W
·
25%L
·
25%T
1085
Elo
43%W
·
29%L
·
28%T
Grok Imagine Image Pro
3 attempts · best result used for ranking
1209
Elo
50%W
·
33%L
·
17%T
1154
Elo
55%W
·
27%L
·
18%T
1147
Elo
33%W
·
44%L
·
23%T
Seedream 4.0
3 attempts · best result used for ranking
1209
Elo
82%W
·
18%L
·
0%T
1156
Elo
50%W
·
40%L
·
10%T
1061
Elo
25%W
·
63%L
·
12%T
Grok Imagine Image
3 attempts · best result used for ranking
1205
Elo
63%W
·
25%L
·
12%T
1158
Elo
45%W
·
45%L
·
10%T
1131
Elo
50%W
·
25%L
·
25%T
Qwen Image Edit 2511
3 attempts · best result used for ranking
1202
Elo
29%W
·
66%L
·
5%T
1042
Elo
9%W
·
91%L
·
0%T
992
Elo
0%W
·
100%L
·
0%T
Nano Banana 2
3 attempts · best result used for ranking
1197
Elo
80%W
·
20%L
·
0%T
1064
Elo
50%W
·
0%L
·
50%T
968
Elo
0%W
·
100%L
·
0%T
Nano Banana
3 attempts · best result used for ranking
1174
Elo
56%W
·
44%L
·
0%T
1167
Elo
55%W
·
18%L
·
27%T
1064
Elo
38%W
·
38%L
·
24%T
Wan 2.6
3 attempts · best result used for ranking
1160
Elo
55%W
·
27%L
·
18%T
1143
Elo
56%W
·
44%L
·
0%T
1061
Elo
25%W
·
75%L
·
0%T
Seedream 4.5
3 attempts · best result used for ranking
1152
Elo
50%W
·
33%L
·
17%T
1131
Elo
44%W
·
56%L
·
0%T
1091
Elo
30%W
·
70%L
·
0%T
FLUX.2 [flex]
3 attempts · best result used for ranking
1145
Elo
38%W
·
54%L
·
8%T
1123
Elo
45%W
·
55%L
·
0%T
1102
Elo
33%W
·
67%L
·
0%T
Z-Image Turbo
3 attempts · best result used for ranking
1034
Elo
8%W
·
92%L
·
0%T
992
Elo
0%W
·
100%L
·
0%T
989
Elo
0%W
·
100%L
·
0%T
Original
Original 100%
Upscaled 100%
×
Hover the original to compare crops at 100%
Drag on the original to compare crops at 100%
RESULT
The most competitive head-to-head matchups, selected by closeness and vote count.
Reve AI
Google
OpenAI
Black Forest Labs
ByteDance
xAI
Alibaba
