Benchmark Almanac

WMT25 EEA Terminology

ONLINE-G 55.9% tops its leaderboard Gemini 2.5 Pro Jun 2025 46.5% leads the models we track
Not in model cards

Models translate English regulatory sentences into Icelandic. The scorer checks whether they use the official Icelandic terms from the European Economic Area agreement.

Passes test 2. The makers publish term accuracy for 34 systems on 256 sentences. A reviewer scored 50 sentences per system. The corrected release contains 299 sentences.

Released
June 2025
Kind
Translation
Level
University
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English, Icelandic
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Run by
Anyone
In views
Law and finance
Strongest model tested
Gemini 2.5 Pro Jun 2025 #97 on the capability index
Last model added
June 2025
Scores refreshed
November 20th, 2026
34
Entries on its own leaderboard, 13 models we track

ONLINE-G tops the leaderboard with 55.9% Gemini 2.5 Pro leads with 46.5% among models released by June 2025

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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 Gemini 2.5 Pro, at 46.5%. This view shows only models we track, each at its best.

The newest model with a score here came out in June 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 ONLINE-G 55.9%
  2. 2 Erlendur 53.3%
  3. 3 ONLINE-B 46.5%
  4. 4 Gemini 2.5 Pro Jun 2025 46.5%
  5. 5 TranssionTranslate 46.3%
  6. 6 hybrid 38.7%
  7. 7 Claude-4 38.5%
  8. 8 SalamandraTA 38.5%
  9. 9 TowerPlus-9B 37.5%
  10. 10 Shy 36.8%
  11. 11 GPT-4.1 34.9%
  12. 12 TowerPlus-72B 32.7%
  13. 13 DeepSeek-V3 Mar 2025 30%
  14. 14 AMI 27.9%
  15. 15 Llama 4 Maverick 27.9%
  16. 16 NLLB 26.6%
  17. 17 CommandA-MT 24.9%
  18. 18 IR-MultiagentMT 24.2%
  19. 19 Gemma 3 27B 20.1%
  20. 20 Mistral Medium 18.9%
  21. 21 GemTrans 16.5%
  22. 22 IRB-MT 13.8%
  23. 23 UvA-MT 13.1%
  24. 24 Gemma 3 12B 11.6%
  25. 25 Qwen3-235B 10.9%
  26. 26 CommandA 7.51%
  27. 27 Llama 3.1-8B 3.15%
  28. 28 Aya Expanse 32B 2.66%
  29. 29 Qwen2.5-7B 0.73%
  30. 30 EuroLLM-9B 0.48%
  31. 31 Command R7B 0.48%
  32. 32 EuroLLM-22B 0.24%
  33. 33 Mistral 7B v0.1 0.24%
  34. 34 Aya Expanse 8B 0%

Source: Official WMT25 EEA Terminology leaderboard

#EntryScoreSettingDateSource
1 ONLINE-G 55.9% Official leaderboard
2 Erlendur 53.3% Official leaderboard
3 ONLINE-B 46.5% Official leaderboard
4 Gemini 2.5 Pro Jun 2025 46.5% Jun 2025 Official leaderboard
5 TranssionTranslate 46.3% Official leaderboard
6 hybrid 38.7% Official leaderboard
7 Claude-4 38.5% Official leaderboard
8 SalamandraTA 38.5% Official leaderboard
9 TowerPlus-9B 37.5% Official leaderboard
10 Shy 36.8% Official leaderboard
11 GPT-4.1 34.9% Apr 2025 Official leaderboard
12 TowerPlus-72B 32.7% Official leaderboard
13 DeepSeek-V3 Mar 2025 30% Mar 2025 Official leaderboard
14 AMI 27.9% Official leaderboard
15 Llama 4 Maverick 27.9% Apr 2025 Official leaderboard
16 NLLB 26.6% Official leaderboard
17 CommandA-MT 24.9% Official leaderboard
18 IR-MultiagentMT 24.2% Official leaderboard
19 Gemma 3 27B 20.1% Mar 2025 Official leaderboard
20 Mistral Medium 18.9% Dec 2023 Official leaderboard
21 GemTrans 16.5% Official leaderboard
22 IRB-MT 13.8% Official leaderboard
23 UvA-MT 13.1% Official leaderboard
24 Gemma 3 12B 11.6% Mar 2025 Official leaderboard
25 Qwen3-235B 10.9% Official leaderboard
26 CommandA 7.51% Official leaderboard
27 Llama 3.1-8B 3.15% Jul 2024 Official leaderboard
28 Aya Expanse 32B 2.66% Oct 2024 Official leaderboard
29 Qwen2.5-7B 0.73% Sep 2024 Official leaderboard
30 EuroLLM-9B 0.48% Official leaderboard
31 Command R7B 0.48% Dec 2024 Official leaderboard
32 EuroLLM-22B 0.24% Official leaderboard
33 Mistral 7B v0.1 0.24% Sep 2023 Official leaderboard
34 Aya Expanse 8B 0% Oct 2024 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 2.5 Pro Jun 2025 46.5% Jun 2025 Official leaderboard
2 GPT-4.1 34.9% Apr 2025 Official leaderboard
3 DeepSeek-V3 Mar 2025 30% Mar 2025 Official leaderboard
4 Llama 4 Maverick 27.9% Apr 2025 Official leaderboard
5 Gemma 3 27B 20.1% Mar 2025 Official leaderboard
6 Mistral Medium 18.9% Dec 2023 Official leaderboard
7 Gemma 3 12B 11.6% Mar 2025 Official leaderboard
8 Llama 3.1-8B 3.15% Jul 2024 Official leaderboard
9 Aya Expanse 32B 2.66% Oct 2024 Official leaderboard
10 Qwen2.5-7B 0.73% Sep 2024 Official leaderboard
11 Command R7B 0.48% Dec 2024 Official leaderboard
12 Mistral 7B v0.1 0.24% Sep 2023 Official leaderboard
13 Aya Expanse 8B 0% Oct 2024 Official leaderboard

2 models have held number one on the official leaderboard since its release in June 2025

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI Google DeepMind
Own leaderboard Not tested since Jun 2025

Source: Official WMT25 EEA Terminology leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Gemini 2.5 Pro GPT-4.1 DeepSeek-V3 Gemma 3 27B Mistral Medium Mistral 7B v0.1 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Gemini 2.5 Pro GPT-4.1

Source: Official WMT25 EEA Terminology 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

All 545 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 17 months ago, in June 2025, for Gemini 2.5 Pro. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (34 entries).
  • Pass Models scored 13 models we track, plus 21 other entries on its leaderboard.
  • Warning Newest score The newest came 17 months ago, in June 2025, for Gemini 2.5 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 2.5 Pro (#97).
  • Pass Headroom The best is 46.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 13 rows for models we track and 21 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.

How we check