Benchmark Almanac

ABACUS

Not in model cards

Models add, remove, or change a specified number of objects in an image. The evaluator checks the count and preservation of the surrounding scene.

Released
October 2026
Kind
Image generation and editing
Level
School
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images → Images
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Scores refreshed
November 20th, 2026
7
Models scored on its own leaderboard

Seedream 4.0 leads with 0.31

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

  1. 1 Seedream 4.0 0.31
  2. 2 Gemini 2.5 Flash Image Preview (nano-banana) 0.29
  3. 3 GPT-Image-1 0.26
  4. 4 Qwen-Image-Edit 0.16
  5. 5 FLUX.1 Kontext [dev] 0.14
  6. 6 Step1X-Edit 0.11
  7. 7 BAGEL 0.09

Source: Official ABACUS leaderboard

#ModelScoreSettingDateSource
1 Seedream 4.0 0.31 Official leaderboard
2 Gemini 2.5 Flash Image Preview (nano-banana) 0.29 Official leaderboard
3 GPT-Image-1 0.26 Official leaderboard
4 Qwen-Image-Edit 0.16 Official leaderboard
5 FLUX.1 Kontext [dev] 0.14 Official leaderboard
6 Step1X-Edit 0.11 Official leaderboard
7 BAGEL 0.09 Official leaderboard

Rankings by part

We read 9 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Add: Easy, Add: Medium and Add: Hard. Seedream 4.0 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Add: Easy

7 models we track have a score here.

  1. 1 Seedream 4.0 0.55
  2. 2 Gemini 2.5 Flash Image Preview (nano-banana) 0.52
  3. 3 GPT-Image-1 0.51
  4. 4 Qwen-Image-Edit 0.35
  5. 5 FLUX.1 Kontext [dev] 0.30
  6. 6 Step1X-Edit 0.24
  7. 7 BAGEL 0.20

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 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 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 official leaderboard (7 entries).
  • Pass Models scored 7 models we track.
  • 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 cea-list.github.io: 7 rows for models we track 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.

How we check