m-ArenaHard
Models answer the 500 Arena-Hard-Auto v0.1 prompts machine-translated into 22 more languages, 23 in all. An LLM judge compares answers pairwise and reports win rates.
Translated with Google Translate API v3. Cohere reports it for Aya Expanse and Command A as win rates against named rivals, so scores depend on the opponent.
- Made by
- Cohere Labs
- Released
- October 2024
- Kind
- Chat and instructions in many languages
- Level
- University
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Last in a system card
- May 2025
- Last model added
- May 2025
2
System cards report it, from 1 lab
No scores here yet
No lab, leaderboard or Epoch AI run has reported a score for m-ArenaHard yet.
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
2 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| May 13, 2025 |  Aya Vision 32B technical report | m-ArenaHard, pairwise win rates averaged over 23 languages, no overall: vs Llama-3.2-90B 56.5, vs Molmo-72B 77.3, vs Qwen-2.5-VL-72B 50.9 (Figure 8, printed labels) | Capabilities | |
| Apr 1, 2025 | Cohere Command A technical report | mArenaHard win rates over 23 languages against single opponents, no absolute score (Table 8): vs Llama 3.3 70B 78.9, Llama 3.1 405B 78.7, Mistral Large 2 65.9, DeepSeek V3 53.4 | Capabilities |
Versions
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.
Nemotron 3 Ultra 550B-A55B Base 90.1% - 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%