UniGenBench

Unified semantic evaluation benchmark for text-to-image generation across style, world knowledge, attributes, actions, relationships, compound prompts, grammar, layout, reasoning, and text.

63rows
overallprimary metric
2026-05-06sampled

Metadata

ID
unigenbench
Category
Multimodal
Release
2025-10-21

Metrics

Overall, Style, World Knowledge, Attribute-Overall, Quantity, Expression, Material, Size, Shape, Color, Action-Overall, Hand, Full body, Animal, Non Contact, Contact, State, Relationship-Overall, Composition, Similarity, Inclusion, Comparison, Compound-Overall, Imagination, Feature matching, Grammar-Overall, Pronoun Reference, Consistency, Negation, Layout-Overall, 2D, 3D, Logical Reasoning, Text

Latest Results

Rows are parsed from the public English UniGenBench leaderboard JSON. Source model display names, links, Hugging Face URLs, and release metadata are preserved.

Rank Subject Overall Model Match Provenance Sampled
1 GPT-4o-1.5 95.77 GPT-4o
openai-gpt-4o
 		Imported 2026-05-06
2 Nano Banana Pro 92.72 Nano Banana Pro (Gemini 3 Pro Image Preview)
google-gemini-3-pro-image-preview
 		Imported 2026-05-06
3 GPT-4o 92.48 GPT-4o
openai-gpt-4o
 		Imported 2026-05-06
4 Imagen-4.0-Ultra-preview-06-06 91.65 — Imported 2026-05-06
5 FLUX-2-max 90.85 — Imported 2026-05-06
6 Seedream-4-5-251128 89.70 — Imported 2026-05-06
7 FLUX-2-flex 89.35 — Imported 2026-05-06
8 FLUX-2-pro 88.35 — Imported 2026-05-06
9 Seedream-4.0 87.35 — Imported 2026-05-06
10 Nano Banana 87.29 Nano Banana (Gemini 2.5 Flash Image)
google-gemini-2.5-flash-image
 		Imported 2026-05-06
11 Imagen-4.0-generate-preview-06-06 85.84 — Imported 2026-05-06
12 FLUX.2-dev 84.76 — Imported 2026-05-06
13 FLUX-kontext-max 80 — Imported 2026-05-06
14 FLUX.2-klein-base-9b 79.35 — Imported 2026-05-06
15 Seedream-3.0 78.41 — Imported 2026-05-06
16 Qwen-Image 78.36 — Imported 2026-05-06
17 FLUX.2-klein-9b 78.28 — Imported 2026-05-06
18 Z-Image 78.10 — Imported 2026-05-06
19 wan2.5-t2i-preview 77.87 — Imported 2026-05-06
20 Imagen-4.0-Fast-preview-06-06 77.69 — Imported 2026-05-06
21 FLUX-kontext-pro 75.84 — Imported 2026-05-06
22 Hunyuan-Image-2.1 74.64 — Imported 2026-05-06
23 LongCat-Image 73.54 — Imported 2026-05-06
24 FLUX.2-klein-4b 72.31 — Imported 2026-05-06
25 Z-Image-Turbo 71.40 — Imported 2026-05-06
26 HiDream-I1-Full 71.36 — Imported 2026-05-06
27 Imagen-3.0-generate-002 71.34 — Imported 2026-05-06
28 Lumina-DiMOO 71.12 — Imported 2026-05-06
29 FLUX-pro-1.1-Ultra 70.46 — Imported 2026-05-06
30 FLUX.1-Krea-dev 69.88 — Imported 2026-05-06
31 FLUX.2-klein-base-4b 69.81 — Imported 2026-05-06
32 Runway-Gen4-Image 69.75 — Imported 2026-05-06
33 Echo-4o 69.12 — Imported 2026-05-06
34 DALL-E-3 68.85 — Imported 2026-05-06
35 Pref-GRPO 68.41 — Imported 2026-05-06
36 GLM-Image 67.23 — Imported 2026-05-06
37 Keling-Ketu 65.23 — Imported 2026-05-06
38 BLIP3-o-Next 65.15 — Imported 2026-05-06
39 wan2.2-t2i-plus 64.82 — Imported 2026-05-06
40 UniWorld-V1 63.11 — Imported 2026-05-06
41 OmniGen2 63.09 — Imported 2026-05-06
42 SD-3.5-Large 62.89 — Imported 2026-05-06
43 Recraft 62.63 — Imported 2026-05-06
44 Stability-AI-stable-image-ultra 61.96 — Imported 2026-05-06
45 Show-o2 61.90 — Imported 2026-05-06
46 HiDream_v2L 61.64 — Imported 2026-05-06
47 Janus-Pro 61.36 — Imported 2026-05-06
48 FLUX.1-dev 60.97 — Imported 2026-05-06
49 SD-3.5-Medium 60.71 — Imported 2026-05-06
50 Bagel 59.91 — Imported 2026-05-06
51 Infinity 59.81 — Imported 2026-05-06
52 BLIP3-o 59.57 — Imported 2026-05-06
53 OneCAT 58.28 — Imported 2026-05-06
54 CogView4 56 — Imported 2026-05-06
55 X-Omni 53.77 — Imported 2026-05-06
56 Janus 51.60 — Imported 2026-05-06
57 Hunyuan-DiT 51.38 — Imported 2026-05-06
58 Janus-flow 47.10 — Imported 2026-05-06
59 Kolors 46.07 — Imported 2026-05-06
60 Playground2.5 46.02 — Imported 2026-05-06
61 Emu3 45.42 — Imported 2026-05-06
62 MMaDA 41.35 — Imported 2026-05-06
63 SDXL 40.22 — Imported 2026-05-06