LingOly
Models infer linguistic rules from examples in unfamiliar languages. They use those rules to solve linguistics olympiad problems.
- Made by
- Andrew M. Bean and colleagues
- Released
- June 2024
- Kind
- Reasoning in many languages
- Level
- Competition
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Subject
- Languages and literature
- Language
- Many languages
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Many languages
- Strongest model tested
- GPT-4o May 2024 #162 on the capability index
- Last model added
- May 2024
- Scores refreshed
- November 20th, 2026
Claude 3 Opus leads with 46.3% among models released by May 2024 Claude 3 Opus leads with 46.3% among models released by May 2024
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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 May 2024, 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 Claude 3 Opus 46.3%
- 2 GPT-4o May 2024 37.6%
- 3 GPT-4 Turbo Apr 2024 33.4%
- 4 Gemini 1.5 Pro (snapshot unspecified) 32.1%
- 5 Command R+ (2024 study) 21.5%
- 6 GPT-3.5 (snapshot unspecified) 21.2%
- 7
Mixtral 8x7B 14.2%
- 8
Llama 3-8B 11.4%
- 9 Aya 23 35B 10.7%
- 10
Llama 3-70B 10.3%
- 11
Llama 2-70B 6.40%
- 12
Gemma 7B 4.90%
Source: The LingOly paper
- 1 Claude 3 Opus 46.3%
- 2 GPT-4o May 2024 37.6%
- 3 GPT-4 Turbo Apr 2024 33.4%
- 4
Mixtral 8x7B 14.2%
- 5
Llama 3-8B 11.4%
- 6
Llama 3-70B 10.3%
- 7
Llama 2-70B 6.40%
- 8
Gemma 7B 4.90%
Source: The LingOly paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3 Opus | 46.3% | Feb 2024 | Official leaderboard | |
| 2 | GPT-4o May 2024 | 37.6% | May 2024 | Official leaderboard | |
| 3 | GPT-4 Turbo Apr 2024 | 33.4% | Apr 2024 | Official leaderboard | |
| 4 | Gemini 1.5 Pro (snapshot unspecified) | 32.1% | Official leaderboard | ||
| 5 | Command R+ (2024 study) | 21.5% | Official leaderboard | ||
| 6 | GPT-3.5 (snapshot unspecified) | 21.2% | Official leaderboard | ||
| 7 | | 14.2% | Dec 2023 | Official leaderboard | |
| 8 | | 11.4% | Apr 2024 | Official leaderboard | |
| 9 | Aya 23 35B | 10.7% | Official leaderboard | ||
| 10 | | 10.3% | Apr 2024 | Official leaderboard | |
| 11 | | 6.40% | Jul 2023 | Official leaderboard | |
| 12 | | 4.90% | Feb 2024 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3 Opus | 46.3% | Feb 2024 | Official leaderboard | |
| 2 | GPT-4o May 2024 | 37.6% | May 2024 | Official leaderboard | |
| 3 | GPT-4 Turbo Apr 2024 | 33.4% | Apr 2024 | Official leaderboard | |
| 4 | | 14.2% | Dec 2023 | Official leaderboard | |
| 5 | | 11.4% | Apr 2024 | Official leaderboard | |
| 6 | | 10.3% | Apr 2024 | Official leaderboard | |
| 7 | | 6.40% | Jul 2023 | Official leaderboard | |
| 8 | | 4.90% | Feb 2024 | Official leaderboard |
Rankings by part
We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as Breakthrough/Foundation, Intermediate/Advanced/Round 2 and Breakthrough/Foundation context gain. Claude 3 Opus leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Breakthrough/Foundation
8 models we track have a score here.
- 1 Claude 3 Opus 77.1%
- 2 GPT-4o May 2024 68.2%
- 3 GPT-4 Turbo Apr 2024 67.3%
- 4
Mixtral 8x7B 28.3%
- 5
Llama 3-8B 26%
- 6
Llama 3-70B 23.8%
- 7
Llama 2-70B 19.3%
- 8
Gemma 7B 7.60%
Intermediate/Advanced/Round 2
8 models we track have a score here.
- 1 Claude 3 Opus 38.7%
- 2 GPT-4o May 2024 30%
- 3 GPT-4 Turbo Apr 2024 25%
- 4
Mixtral 8x7B 10.7%
- 5
Llama 3-8B 7.80%
- 6
Llama 3-70B 6.90%
- 7
Gemma 7B ↑1 4.20%
- 8
Llama 2-70B ↓1 3.20%
Breakthrough/Foundation context gain
8 models we track have a score here.
- 1 Claude 3 Opus 45.3%
- 2 GPT-4 Turbo Apr 2024 ↑1 40.8%
- 3 GPT-4o May 2024 ↓1 38.6%
- 4
Llama 3-8B ↑1 11.7%
- 5
Mixtral 8x7B ↓1 9.00%
- 6
Llama 2-70B ↑1 8.10%
- 7
Llama 3-70B ↓1 7.60%
- 8
Gemma 7B 2.20%
Intermediate/Advanced/Round 2 context gain
8 models we track have a score here.
- 1 Claude 3 Opus 24.7%
- 2 GPT-4o May 2024 21.7%
- 3 GPT-4 Turbo Apr 2024 16.7%
- 4
Mixtral 8x7B 5.80%
- 5
Llama 3-8B 3.20%
- 6
Gemma 7B ↑2 2.20%
- 7
Llama 3-70B ↓1 1.80%
- 8
Llama 2-70B ↓1 -0.70%
Overall context gain
8 models we track have a score here.
- 1 Claude 3 Opus 28.8%
- 2 GPT-4o May 2024 25.1%
- 3 GPT-4 Turbo Apr 2024 21.5%
- 4
Mixtral 8x7B 6.40%
- 5
Llama 3-8B 4.90%
- 6
Llama 3-70B 2.90%
- 7
Gemma 7B ↑1 2.20%
- 8
Llama 2-70B ↓1 1.10%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The LingOly 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 LingOly 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
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in May 2024, for GPT-4o. 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 benchmark’s paper (12 entries).
- Pass Models scored 8 models we track, plus 4 other entries on its leaderboard.
- Warning Newest score The newest came 2 years ago, in May 2024, for GPT-4o.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-4o (#162).
- Pass Headroom The best is 46.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 8 rows for models we track and 4 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.