Benchmark Almanac

Rapidata Multi-Step Image Editing (Identity Drift)

Ideogram 4.6 (Precise Edit, High Quality) 1,892 tops its leaderboard Ideogram 4.5 1,866 leads the models we track
Not in model cards

Image editors apply five sequential changes to 199 photographs, and people compare the final faces for identity drift.

The benchmark ranks 23 models with Bradley-Terry ratings from 413,667 human responses across 76,714 matches. Rapidata publishes the prompt chains, model outputs, and evaluation results under CC BY-SA 4.0, and Ideogram's 4.6 page reports the November 17 leaderboard snapshot.

Made by
Rapidata
Released
November 2026
Kind
Image generation and editing
Level
Expert
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Measures
Preference
Language
English
In and out
Images, Text → Images
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Last model added
November 2026
Scores refreshed
November 21st, 2026
12
Entries on its own leaderboard, 6 models we track

Ideogram 4.6 (Precise Edit, High Quality) tops the leaderboard with 1,892 Ideogram 4.5 leads with 1,866

Best score per entry from the benchmark's own leaderboard, read on November 21st, 2026. 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 Ideogram 4.5, at 1,866. This view shows only models we track, each at its best.

Bars start at 800.

  1. 1 Ideogram 4.6 (Precise Edit, High Quality) 1,892
  2. 2 Ideogram 4.5 high quality, high precision 1,866
  3. 3 Seedream 5.5 1,805
  4. 4 FLUX 3 Image 1,697
  5. 5 Muse Image 1,406
  6. 6 MAI-Image-2.6 1,299
  7. 7 Qwen-Image-2.1-Pro (Thinking) 1,291
  8. 8 Seedream 5.0 Pro 1,174
  9. 9 MAI-Image-2.5-Pro 1,145
  10. 10 Qwen-Image-3.0-Pro thinking 1,014
  11. 11 GPT-Image-2.6 Sunburst max quality, high 1,001
  12. 12 Nano Banana Pro 987

Source: Official Rapidata Multi-Step Image Editing (Identity Drift) leaderboard

#EntryScore ±SettingSubmitted byDateSource
1 Ideogram 4.6 (Precise Edit, High Quality) 1,892 ±32 Ideogram Official leaderboard
2 Ideogram 4.5 1,866 ±30 high quality, high precision Ideogram Official leaderboard
3 Seedream 5.5 1,805 ±29 ByteDance Official leaderboard
4 FLUX 3 Image 1,697 ±27 Black Forest Labs Official leaderboard
5 Muse Image 1,406 ±22 Meta Official leaderboard
6 MAI-Image-2.6 1,299 ±22 Microsoft Official leaderboard
7 Qwen-Image-2.1-Pro (Thinking) 1,291 ±21 Alibaba Cloud Official leaderboard
8 Seedream 5.0 Pro 1,174 ±21 ByteDance Official leaderboard
9 MAI-Image-2.5-Pro 1,145 ±21 Microsoft Official leaderboard
10 Qwen-Image-3.0-Pro 1,014 ±21 thinking Alibaba Cloud Official leaderboard
11 GPT-Image-2.6 Sunburst 1,001 ±21 max quality, high OpenAI Nov 2026 Official leaderboard
12 Nano Banana Pro 987 ±21 Google Official leaderboard
#ModelScore ±SettingSubmitted byDateSource
1 Ideogram 4.5 1,866 ±30 high quality, high precision Ideogram Official leaderboard
2 Muse Image 1,406 ±22 Meta Official leaderboard
3 MAI-Image-2.6 1,299 ±22 Microsoft Official leaderboard
4 Seedream 5.0 Pro 1,174 ±21 ByteDance Official leaderboard
5 Qwen-Image-3.0-Pro 1,014 ±21 thinking Alibaba Cloud Official leaderboard
6 GPT-Image-2.6 Sunburst 1,001 ±21 max quality, high OpenAI Nov 2026 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
1,001 2027 1,001 2027

Source: Official Rapidata Multi-Step Image Editing (Identity Drift) 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

All 636 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

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (12 entries).
  • Pass Models scored 6 models we track, plus 6 other entries on its leaderboard.
  • Pass Newest score The newest is from November 2026, for GPT-Image-2.6 Sunburst.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Scored as an Elo rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on benchmark.ai: 6 rows for models we track and 6 other entries at the last read.
  • Pass Last read November 21st, 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.

How we check