MMArt-Bench
Models retouch 200 photos from editing instructions. Gemini scores visual quality and instruction following, and image comparisons measure how closely edits match reference photos.
The EyeControl team independently tested local edits on the portrait subset. The original study also evaluates edits across three other photographic genres.
- Made by
- Xiamen University and collaborators
- Released
- June 2025
- Announced on X
- 5.8K views, 31 likes
- Kind
- Apps and MCP servers
- Level
- University
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Subject
- Arts
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision, Image, video and 3D generation
- Scores refreshed
- November 20th, 2026
GPT-4o (snapshot unspecified) tops the leaderboard with 9.11 Step1X-Edit leads with 8.34
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The headline averages the overall VIEScores for scene-level and region-level edits. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Step1X-Edit, at 8.34. This view shows only models we track, each at its best.
- 1 GPT-4o (snapshot unspecified) Gemini-2-Flash judge 9.11
- 2 JarvisArt Gemini-2-Flash judge 8.61
- 3 Gemini-2-Flash (image editing, snapshot unspecified) Gemini-2-Flash judge 8.35
- 4
Step1X-Edit Gemini-2-Flash judge 8.34
- 5 InstructPix2Pix Gemini-2-Flash judge 6.00
- 6 OmniGen Gemini-2-Flash judge 5.42
- 7 MagicBrush Gemini-2-Flash judge 3.49
- 8 VARGPT-v1.1 Gemini-2-Flash judge 1.28
Source: The MMArt-Bench paper
Source: The MMArt-Bench paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o (snapshot unspecified) | 9.11 | Gemini-2-Flash judge | Official leaderboard | |
| 2 | JarvisArt | 8.61 | Gemini-2-Flash judge | Official leaderboard | |
| 3 | Gemini-2-Flash (image editing, snapshot unspecified) | 8.35 | Gemini-2-Flash judge | Official leaderboard | |
| 4 | | 8.34 | Gemini-2-Flash judge | Official leaderboard | |
| 5 | InstructPix2Pix | 6.00 | Gemini-2-Flash judge | Official leaderboard | |
| 6 | OmniGen | 5.42 | Gemini-2-Flash judge | Official leaderboard | |
| 7 | MagicBrush | 3.49 | Gemini-2-Flash judge | Official leaderboard | |
| 8 | VARGPT-v1.1 | 1.28 | Gemini-2-Flash judge | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 8.34 | Gemini-2-Flash judge | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
No dated values yet.
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 Tool use
- AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state.
Gemini 4 Neon 56.3%
- MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks.
Gemini 4 Argon 90.7%
- BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes.
Qwen3.7-Max 75%
- τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state.
Qwen 3.8 Max 55.1%
- Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
- DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%
Data audit
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (8 entries).
- Pass Models scored 1 model we track, plus 7 other entries on its leaderboard.
- Note Newest score No scores yet.
- 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: 2 rows for models we track and 7 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.