ImgMT
Models translate text embedded in diagrams, maps, and infographics. Evaluation measures text recognition, region matching, and translation with CER, F1, chrF, and COMET.
The dataset contains 35,612 aligned image pairs in 129 languages. Report the test split and metric because error rates and quality scores have opposite directions.
- Made by
- Šárka Uramová, Petar Petrov, Josef Jon, Ondřej Bojar, and Michal Novák
- Released
- October 2026
- Kind
- Translation
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Subject
- Languages and literature
- Language
- Many languages
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision, Many languages
- Strongest model tested
- Kimi K3 #25 on the capability index
- Last model added
- July 2026
- Scores refreshed
- November 20th, 2026
In the authors’ words
Text-in-image translation aims to translate all textual content embedded in an image. In Text Image Translation (TIT), the output is translated text; In-Image Machine Translation (IIMT) additionally renders the translation back into the image. We introduce ImgMT, a massively multilingual dataset and benchmark mined from localized SVG illustrations on Wikimedia Commons, focusing on diagrams, maps, and infographics. ImgMT contains 35,612 aligned image pairs grouped into 1,855 image sets, spanning 129 languages, 20+ writing systems, and 4,886 language pairs. We propose a multi-level evaluation protocol that measures source-side text-region detection and OCR accuracy alongside target-side translation quality. We benchmark several multimodal large language models and a dedicated TIT system in English-to-X and X-to-English settings, and find that the source language and writing system substantially affect TIT quality. Finally, we show that ImgMT is not well suited for isolating the contribution of visual information in multimodal translation.
Kimi K3 leads with 61.3 among models released by July 2026 Kimi K3 leads with 61.3 among models released by July 2026
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports multiple measures, so this headline is chrF. 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. This view shows only models we track, each at its best.
The newest model with a score here came out in July 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
- 1 Kimi K3 61.3
- 2 Qwen3.5 59.9
- 3 GPT-5.6 Luna 59.7
- 4 AnyTrans 57.5
- 5 Claude Haiku 4.5 39.8
Source: The ImgMT paper
- 1 Kimi K3 61.3
- 2 GPT-5.6 Luna 59.7
- 3 Claude Haiku 4.5 39.8
Source: The ImgMT paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Kimi K3 | 61.3 | Jul 2026 | Official leaderboard | |
| 2 | Qwen3.5 | 59.9 | Official leaderboard | ||
| 3 | GPT-5.6 Luna | 59.7 | Jul 2026 | Official leaderboard | |
| 4 | AnyTrans | 57.5 | Official leaderboard | ||
| 5 | Claude Haiku 4.5 | 39.8 | Oct 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Kimi K3 | 61.3 | Jul 2026 | Official leaderboard | |
| 2 | GPT-5.6 Luna | 59.7 | Jul 2026 | Official leaderboard | |
| 3 | Claude Haiku 4.5 | 39.8 | Oct 2025 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s paper, each ranking models on one part of the test, such as COMET, CER and BBA. GPT-5.6 Luna leads 2 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.
COMET
3 models we track have a score here.
- 1 Kimi K3 83.4
- 2 GPT-5.6 Luna 82.6
- 3 Claude Haiku 4.5 68.1
CER
3 models we track have a score here.
- 1 Claude Haiku 4.5 ↑2 0.82
- 2 GPT-5.6 Luna 0.22
- 3 Kimi K3 ↓2 0.17
BBA
3 models we track have a score here.
- 1 GPT-5.6 Luna ↑1 0.66
- 2 Kimi K3 ↓1 0.62
- 3 Claude Haiku 4.5 0.20
BBC
3 models we track have a score here.
- 1 GPT-5.6 Luna ↑1 0.84
- 2 Kimi K3 ↓1 0.83
- 3 Claude Haiku 4.5 0.54
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The ImgMT 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 ImgMT 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 Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
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 benchmark’s paper (5 entries).
- Pass Models scored 3 models we track, plus 2 other entries on its leaderboard.
- Pass Newest score The newest is from July 2026, for Kimi K3.
- Note Frontier models None of the top 10 has a score. The strongest tested is Kimi K3 (#25).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on aclanthology.org: 3 rows for models we track and 2 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.