Model leaderboard

Every model we’ve run through the standard Glif tests, scored out of 10 by an LLM judge. Overall needs 3 judged tests to rank; pick a capability to rank on it alone.

1
GPT Image 2
3 tests
8.4
2
Nano Banana Pro Image Generator
10 tests
8.2
3
FLUX.2 Max Image Generator
11 tests
8.0
4
FLUX.2 Pro Image Generator
11 tests
7.9
5
FLUX.2 Turbo Image Generator
10 tests
7.8
6
FLUX.2 Flex Image Generator
11 tests
7.7
7
Nano Banana Image Generator
6 tests
7.7
8
Seedream V4.5
10 tests
7.6
9
GPT Image 1.5
19 tests
7.5
10
Krea 2 Image Generator
9 tests
7.3
11
FLUX.1 Dev Image Generator
11 tests
7.1
12
Image to Video (Kling O3 Pro)
5 tests
7.0
13
Image to Video (Kling V3 Pro)
3 tests
7.0
14
Image to Video (Kling V3 Standard)
3 tests
7.0
15
Image to Video (Kling 2.5 Pro)
3 tests
6.9
16
Image to Video (Kling 2.6 Pro)
4 tests
6.9
17
FLUX.2 Klein 9B Image Generator
11 tests
6.9
18
Image to Video (Vidu Q3)
4 tests
6.8
19
Image to Video (Veo 3.1)
3 tests
6.7
20
FLUX.1 Schnell Image Generator
11 tests
6.7
21
Qwen Image Generator with LoRA
16 tests
6.5
22
Grok Imagine Video
5 tests
6.4
23
Clarity Upscaler
3 tests
6.3
24
Add Video Subtitles (VEED)
4 tests
6.3
25
Reference to Video (Kling O3 Pro)
5 tests
6.2
26
Image to Video (Veo 3.1 Fast)
3 tests
6.0
27
Video (Veo 3.1 Lite)
3 tests
6.0
28
Seedance Pro 1.5
5 tests
5.9
29
Sora 2
4 tests
5.5
30
Z-Image Generator
4 tests
5.4
31
FLUX Kontext Image Editor
5 tests
5.3
32
P-Image Upscale
3 tests
4.8
33
Recraft Crisp Upscaler
3 tests
3.7
34
Ideogram Layerize Text
4 tests
2.4