Okapi MMLU
Qwen2.5-72B 80% leads since September 2024 Models answer MMLU multiple choice questions that ChatGPT translated into 26 languages. Scoring is accuracy.
The questions are machine translated, so translation errors and English cultural bias remain.
- Released
- July 2023
- Kind
- Knowledge in many languages
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Strongest model tested
- Qwen2.5-72B #165 on the capability index
- Last in a launch post
- September 2024, dropped before it saturated
- Last model added
- July 2025
Qwen2.5-72B leads with 80% among models released by May 2025
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in May 2025, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 16 more
- 1 Qwen2.5-72B 80%
- 2 Claude 3 Opus 79.1%
- 3
Falcon-H1-34B 77.8% - 4 Qwen2.5-32B 77.2%
- 5 Qwen2.5-14B 72.1%
- 6 Claude 3 Sonnet 69%
- 7 Qwen2.5-7B 67%
- 8 Claude 3 Haiku 65.2%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen2.5-72B | 80% | Sep 2024 | Lab model card | |
| 2 | Claude 3 Opus | 79.1% | Mar 2024 | Lab model card | |
| 3 | Falcon-H1-34B | 77.8% | Jul 2025 | Lab model card | |
| 4 | Qwen2.5-32B | 77.2% | Dec 2024 | Lab model card | |
| 5 | Qwen2.5-14B | 72.1% | Dec 2024 | Lab model card | |
| 6 | Claude 3 Sonnet | 69% | Mar 2024 | Lab model card | |
| 7 | Qwen2.5-7B | 67% | Dec 2024 | Lab model card | |
| 8 | Claude 3 Haiku | 65.2% | Mar 2024 | Lab model card |
2 models have held number one in labs’ own numbers since its release in July 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Labs’ model cards
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Labs’ model cards
No launch post has listed it since September 2024
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Sep 19, 2024 | Qwen2.5-72B | 80 | okapi MMLU (translated) |
8 system cards report it, from 3 labs
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 |
|---|---|---|---|---|
| Jul 30, 2025 | Falcon-H1-34B technical report | 77.8 | Multi-MMLU (Dac Lai et al.), 12 languages, logprobs 5-shot (Tables 24 and 35) | Capabilities |
| Dec 19, 2024 | Qwen2.5-7B technical report | 67 | okapi MMLU (translated), Qwen2.5-7B-Instruct (Table 14) | Capabilities |
| Dec 19, 2024 | Qwen2.5-72B technical report | 80 | okapi MMLU (translated), Qwen2.5-72B-Instruct (Table 13) | Capabilities |
| Dec 19, 2024 | Qwen2.5-32B technical report | 77.2 | okapi MMLU (translated), Qwen2.5-32B-Instruct (Table 13) | Capabilities |
| Dec 19, 2024 | Qwen2.5-14B technical report | 72.1 | okapi MMLU (translated), Qwen2.5-14B-Instruct (Table 14) | Capabilities |
| Mar 4, 2024 | Claude 3 Sonnet model card | 69 | Multilingual MMLU (Okapi translation), 5-shot. Per-language chart without printed values (Table 5, Figure 10) | Capabilities |
| Mar 4, 2024 | Claude 3 Opus model card | 79.1 | Multilingual MMLU (Okapi translation), 5-shot. Per-language chart without printed values (Table 5, Figure 10) | Capabilities |
| Mar 4, 2024 | Claude 3 Haiku model card | 65.2 | Multilingual MMLU (Okapi translation), 5-shot. Per-language chart without printed values (Table 5, Figure 10) | Capabilities |
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%