MoI-Bench
Models generate and edit images for mobile users. The authors compare image quality across 64 subtasks and measure the cost of running models on phones.
- Made by
- Kailai Feng and colleagues
- Released
- October 2026
- Kind
- Image generation and editing
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision, Image, video and 3D generation
- Last model added
- February 2026
- Scores refreshed
- November 20th, 2026
GPT-Image-2 leads with 0.97 among models released by February 2026 GPT-Image-2 leads with 0.97 among models released by February 2026
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 0.97 GPT-Image-2
- 0.95
Seedream 5.0 Pro
- 0.95
Gemini 3.1 Flash Image
- 0.83 FLUX.2 [klein] 4B
- 0.69 DeepGen 1.0
- 0.67 Lance-3B
- 0.67 LongCat-Image-Edit
- 0.66 Ovis-U1-3B
- 0.64
BAGEL-7B-MoT
- 0.62 DreamLite
- 0.62 OmniGen2
- 0.59 VIBE-Image-Edit
- 0.51 InternVL-U-4B
- 0.43 Lumina-DiMOO-8B
- 0.12 LongCat-Image
- 0.10 Mobile-O-0.5B
- 0.10 SANA-600M
- 0.10 SANA-Sprint-0.6B
- 0.08 Janus-Pro-1B
- 1 GPT-Image-2 GPT-4o judge 0.97
- 2
Seedream 5.0 Pro GPT-4o judge 0.95
- 3
Gemini 3.1 Flash Image GPT-4o judge 0.95
- 4 FLUX.2 [klein] 4B GPT-4o judge 0.83
- 5 DeepGen 1.0 GPT-4o judge 0.69
- 6 Lance-3B GPT-4o judge 0.67
- 7 LongCat-Image-Edit GPT-4o judge 0.67
- 8 Ovis-U1-3B GPT-4o judge 0.66
- 9
BAGEL-7B-MoT GPT-4o judge 0.64
- 10 DreamLite GPT-4o judge 0.62
- 11 OmniGen2 GPT-4o judge 0.62
- 12 VIBE-Image-Edit GPT-4o judge 0.59
- 13 InternVL-U-4B GPT-4o judge 0.51
- 14 Lumina-DiMOO-8B GPT-4o judge 0.43
- 15 LongCat-Image GPT-4o judge 0.12
- 16 Mobile-O-0.5B GPT-4o judge 0.10
- 17 SANA-600M GPT-4o judge 0.10
- 18 SANA-Sprint-0.6B GPT-4o judge 0.10
- 19 Janus-Pro-1B GPT-4o judge 0.08
Source: The MoI-Bench paper
- 1 GPT-Image-2 GPT-4o judge 0.97
- 2
Seedream 5.0 Pro GPT-4o judge 0.95
- 3
Gemini 3.1 Flash Image GPT-4o judge 0.95
- 4 FLUX.2 [klein] 4B GPT-4o judge 0.83
- 5
BAGEL-7B-MoT GPT-4o judge 0.64
Source: The MoI-Bench paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-Image-2 | 0.97 | GPT-4o judge | Official leaderboard | |
| 2 | | 0.95 | GPT-4o judge | Official leaderboard | |
| 3 | | 0.95 | GPT-4o judge | Feb 2026 | Official leaderboard |
| 4 | FLUX.2 [klein] 4B | 0.83 | GPT-4o judge | Official leaderboard | |
| 5 | DeepGen 1.0 | 0.69 | GPT-4o judge | Official leaderboard | |
| 6 | Lance-3B | 0.67 | GPT-4o judge | Official leaderboard | |
| 7 | LongCat-Image-Edit | 0.67 | GPT-4o judge | Official leaderboard | |
| 8 | Ovis-U1-3B | 0.66 | GPT-4o judge | Official leaderboard | |
| 9 | | 0.64 | GPT-4o judge | May 2025 | Official leaderboard |
| 10 | DreamLite | 0.62 | GPT-4o judge | Official leaderboard | |
| 11 | OmniGen2 | 0.62 | GPT-4o judge | Official leaderboard | |
| 12 | VIBE-Image-Edit | 0.59 | GPT-4o judge | Official leaderboard | |
| 13 | InternVL-U-4B | 0.51 | GPT-4o judge | Official leaderboard | |
| 14 | Lumina-DiMOO-8B | 0.43 | GPT-4o judge | Official leaderboard | |
| 15 | LongCat-Image | 0.12 | GPT-4o judge | Official leaderboard | |
| 16 | Mobile-O-0.5B | 0.10 | GPT-4o judge | Official leaderboard | |
| 17 | SANA-600M | 0.10 | GPT-4o judge | Official leaderboard | |
| 18 | SANA-Sprint-0.6B | 0.10 | GPT-4o judge | Official leaderboard | |
| 19 | Janus-Pro-1B | 0.08 | GPT-4o judge | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-Image-2 | 0.97 | GPT-4o judge | Official leaderboard | |
| 2 | | 0.95 | GPT-4o judge | Official leaderboard | |
| 3 | | 0.95 | GPT-4o judge | Feb 2026 | Official leaderboard |
| 4 | FLUX.2 [klein] 4B | 0.83 | GPT-4o judge | Official leaderboard | |
| 5 | | 0.64 | GPT-4o judge | May 2025 | Official leaderboard |
Rankings by part
We read 47 parts from the benchmark’s paper in 16 groups: criterion group, generation, local editing, global editing, understand. editing, ref. editing, general objects, animals, portraits, text&docs., if, vq, dp, operation, target and criteria. Each ranks models on one part of the test. GPT-Image-2 leads 41 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Instruction following
5 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.95
- 4 FLUX.2 [klein] 4B 0.82
- 5
BAGEL-7B-MoT 0.69
Visual quality
5 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Seedream 5.0 Pro 0.97
- 3
Gemini 3.1 Flash Image 0.97
- 4 FLUX.2 [klein] 4B 0.88
- 5
BAGEL-7B-MoT 0.68
Detail preservation
5 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Seedream 5.0 Pro 0.98
- 3
Gemini 3.1 Flash Image 0.97
- 4 FLUX.2 [klein] 4B 0.95
- 5
BAGEL-7B-MoT 0.77
T2I
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Gemini 3.1 Flash Image ↑1 0.99
- 3
Seedream 5.0 Pro ↓1 0.98
- 4 FLUX.2 [klein] 4B 0.90
Add.
4 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.82
Rem.
4 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.96
- 3
Gemini 3.1 Flash Image 0.95
- 4 FLUX.2 [klein] 4B 0.85
Repl.
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Seedream 5.0 Pro 0.98
- 3
Gemini 3.1 Flash Image 0.98
- 4 FLUX.2 [klein] 4B 0.87
Adj.
4 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Seedream 5.0 Pro 0.96
- 3
Gemini 3.1 Flash Image 0.96
- 4 FLUX.2 [klein] 4B 0.84
Style
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Gemini 3.1 Flash Image ↑1 0.99
- 3 FLUX.2 [klein] 4B ↑1 0.97
- 4
Seedream 5.0 Pro ↓2 0.94
Bkg.
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2 FLUX.2 [klein] 4B ↑2 0.97
- 3
Seedream 5.0 Pro ↓1 0.97
- 4
Gemini 3.1 Flash Image ↓1 0.97
Rest.
4 models we track have a score here.
- 1 GPT-Image-2 0.95
- 2
Seedream 5.0 Pro 0.92
- 3
Gemini 3.1 Flash Image 0.92
- 4 FLUX.2 [klein] 4B 0.85
View
4 models we track have a score here.
- 1
Seedream 5.0 Pro ↑1 0.83
- 2 GPT-Image-2 ↓1 0.82
- 3
Gemini 3.1 Flash Image 0.80
- 4 FLUX.2 [klein] 4B 0.78
Sem.
4 models we track have a score here.
- 1 GPT-Image-2 0.95
- 2
Gemini 3.1 Flash Image ↑1 0.95
- 3
Seedream 5.0 Pro ↓1 0.93
- 4 FLUX.2 [klein] 4B 0.57
Outp.
4 models we track have a score here.
- 1
Gemini 3.1 Flash Image ↑2 0.95
- 2 GPT-Image-2 ↓1 0.92
- 3
Seedream 5.0 Pro ↓1 0.87
- 4 FLUX.2 [klein] 4B 0.80
Inp.
4 models we track have a score here.
- 1 GPT-Image-2 1.00
- 2
Seedream 5.0 Pro 1.00
- 3 FLUX.2 [klein] 4B ↑1 0.99
- 4
Gemini 3.1 Flash Image ↓1 0.99
R1
4 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.97
- 3
Gemini 3.1 Flash Image 0.96
- 4 FLUX.2 [klein] 4B 0.78
B1+R1
4 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Seedream 5.0 Pro 0.94
- 3
Gemini 3.1 Flash Image 0.93
- 4 FLUX.2 [klein] 4B 0.71
B1+R2
4 models we track have a score here.
- 1
Seedream 5.0 Pro ↑1 0.90
- 2
Gemini 3.1 Flash Image ↑1 0.88
- 3 GPT-Image-2 ↓2 0.87
- 4 FLUX.2 [klein] 4B 0.61
B1+R3
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Seedream 5.0 Pro 0.99
- 3
Gemini 3.1 Flash Image 0.80
- 4 FLUX.2 [klein] 4B 0.70
Concrete
4 models we track have a score here.
- 1 GPT-Image-2 0.95
- 2
Gemini 3.1 Flash Image ↑1 0.95
- 3
Seedream 5.0 Pro ↓1 0.93
- 4 FLUX.2 [klein] 4B 0.87
Abstract
4 models we track have a score here.
- 1
Seedream 5.0 Pro ↑1 0.98
- 2 GPT-Image-2 ↓1 0.97
- 3
Gemini 3.1 Flash Image 0.95
- 4 FLUX.2 [klein] 4B 0.85
Animals
4 models we track have a score here.
- 1
Gemini 3.1 Flash Image ↑2 0.99
- 2 GPT-Image-2 ↓1 0.98
- 3
Seedream 5.0 Pro ↓1 0.96
- 4 FLUX.2 [klein] 4B 0.89
Body&Limbs
4 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.83
Face&Hair
4 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Seedream 5.0 Pro 0.96
- 3
Gemini 3.1 Flash Image 0.95
- 4 FLUX.2 [klein] 4B 0.87
Cloth.&Accs.
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Seedream 5.0 Pro 0.99
- 3
Gemini 3.1 Flash Image 0.97
- 4 FLUX.2 [klein] 4B 0.92
Multi-subjects
4 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.80
Text&Docs.
4 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Seedream 5.0 Pro 0.94
- 3
Gemini 3.1 Flash Image 0.93
- 4 FLUX.2 [klein] 4B 0.62
Fidelity
4 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Seedream 5.0 Pro 0.94
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.81
Selection
4 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.96
- 3
Gemini 3.1 Flash Image 0.95
- 4 FLUX.2 [klein] 4B 0.80
Integrity
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Seedream 5.0 Pro 0.98
- 3
Gemini 3.1 Flash Image 0.98
- 4 FLUX.2 [klein] 4B 0.88
Coherence
4 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.96
- 3
Gemini 3.1 Flash Image 0.96
- 4 FLUX.2 [klein] 4B 0.87
Presentation
4 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Seedream 5.0 Pro 0.98
- 3
Gemini 3.1 Flash Image 0.98
- 4 FLUX.2 [klein] 4B 0.90
Context
4 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Gemini 3.1 Flash Image ↑1 0.98
- 3
Seedream 5.0 Pro ↓1 0.97
- 4 FLUX.2 [klein] 4B 0.95
Structure
4 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Seedream 5.0 Pro 0.98
- 3 FLUX.2 [klein] 4B ↑1 0.96
- 4
Gemini 3.1 Flash Image ↓1 0.96
Identity
4 models we track have a score here.
- 1
Gemini 3.1 Flash Image ↑2 0.99
- 2 GPT-Image-2 ↓1 0.98
- 3
Seedream 5.0 Pro ↓1 0.98
- 4 FLUX.2 [klein] 4B 0.94
Gen.
5 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Gemini 3.1 Flash Image ↑1 0.97
- 3
Seedream 5.0 Pro ↓1 0.97
- 4 FLUX.2 [klein] 4B 0.90
- 5
BAGEL-7B-MoT 0.84
Local
5 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.80
- 5
BAGEL-7B-MoT 0.70
Global
5 models we track have a score here.
- 1 GPT-Image-2 0.93
- 2
Gemini 3.1 Flash Image ↑1 0.89
- 3
Seedream 5.0 Pro ↓1 0.86
- 4 FLUX.2 [klein] 4B 0.81
- 5
BAGEL-7B-MoT 0.53
Underst.
5 models we track have a score here.
- 1 GPT-Image-2 0.94
- 2
Seedream 5.0 Pro 0.91
- 3
Gemini 3.1 Flash Image 0.89
- 4 FLUX.2 [klein] 4B 0.67
- 5
BAGEL-7B-MoT 0.47
Ref.
5 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.70
- 5
BAGEL-7B-MoT 0.14
General
5 models we track have a score here.
- 1 GPT-Image-2 0.95
- 2
Gemini 3.1 Flash Image ↑1 0.93
- 3
Seedream 5.0 Pro ↓1 0.92
- 4 FLUX.2 [klein] 4B 0.81
- 5
BAGEL-7B-MoT 0.58
Animals
5 models we track have a score here.
- 1 GPT-Image-2 0.99
- 2
Gemini 3.1 Flash Image ↑1 0.97
- 3
Seedream 5.0 Pro ↓1 0.95
- 4 FLUX.2 [klein] 4B 0.90
- 5
BAGEL-7B-MoT 0.72
Portraits
5 models we track have a score here.
- 1 GPT-Image-2 0.97
- 2
Seedream 5.0 Pro 0.95
- 3
Gemini 3.1 Flash Image 0.94
- 4 FLUX.2 [klein] 4B 0.81
- 5
BAGEL-7B-MoT 0.64
Text.
5 models we track have a score here.
- 1 GPT-Image-2 0.93
- 2
Seedream 5.0 Pro 0.90
- 3
Gemini 3.1 Flash Image 0.90
- 4 FLUX.2 [klein] 4B 0.60
- 5
BAGEL-7B-MoT 0.44
IF
5 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Gemini 3.1 Flash Image ↑1 0.94
- 3
Seedream 5.0 Pro ↓1 0.93
- 4 FLUX.2 [klein] 4B 0.79
- 5
BAGEL-7B-MoT 0.67
VQ
5 models we track have a score here.
- 1 GPT-Image-2 0.98
- 2
Seedream 5.0 Pro 0.97
- 3
Gemini 3.1 Flash Image 0.96
- 4 FLUX.2 [klein] 4B 0.85
- 5
BAGEL-7B-MoT 0.64
DP
5 models we track have a score here.
- 1 GPT-Image-2 0.96
- 2
Gemini 3.1 Flash Image ↑1 0.96
- 3
Seedream 5.0 Pro ↓1 0.96
- 4 FLUX.2 [klein] 4B 0.93
- 5
BAGEL-7B-MoT 0.77
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The MoI-Bench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-Image-2 | GPT-4o judge | 0.97 | |||
| GPT-Image-2 | Gemini 3.1 Pro judge | 0.96 | |||
| GPT-4o judge | 0.95 | ||||
| GPT-4o judge | 0.95 | ||||
| Gemini 3.1 Pro judge | 0.93 | ||||
| Gemini 3.1 Pro judge | 0.93 | ||||
| FLUX.2 [klein] 4B | GPT-4o judge | 0.83 | |||
| FLUX.2 [klein] 4B | Gemini 3.1 Pro judge | 0.79 | |||
| GPT-4o judge | 0.64 | ||||
| Gemini 3.1 Pro judge | 0.60 |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The MoI-Bench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-Image-2 | GPT-4o judge | 0.97 | |||
| GPT-Image-2 | Gemini 3.1 Pro judge | 0.96 | |||
| GPT-4o judge | 0.95 | ||||
| GPT-4o judge | 0.95 | ||||
| Gemini 3.1 Pro judge | 0.93 | ||||
| Gemini 3.1 Pro judge | 0.93 | ||||
| FLUX.2 [klein] 4B | GPT-4o judge | 0.83 | |||
| FLUX.2 [klein] 4B | Gemini 3.1 Pro judge | 0.79 | |||
| GPT-4o judge | 0.64 | ||||
| Gemini 3.1 Pro judge | 0.60 |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The MoI-Bench paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official MoI-Bench leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Gemini 3.1 Flash Image. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (19 entries).
- Pass Models scored 5 models we track, plus 14 other entries on its leaderboard.
- Warning Newest score The newest came 9 months ago, in February 2026, for Gemini 3.1 Flash Image.
- Note Frontier models None of its models are on the capabilities index.
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 10 rows for models we track and 14 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.