BertaQA
Models answer questions about Basque culture and general knowledge. Parallel Basque and English questions help distinguish language effects from familiarity with local topics.
- Made by
- Julen Etxaniz and colleagues
- Released
- June 2024
- Kind
- Trivia and open questions
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- Claude 3 Opus #173 on the capability index
- Last model added
- June 2024
- Scores refreshed
- November 20th, 2026
Claude 3 Opus leads with 81.5% among models released by June 2024 Claude 3 Opus leads with 81.5% among models released by June 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 June 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
- 81.5 Claude 3 Opus
- 80.7 GPT-4 Turbo (Jan 2024)
- 77.6 GPT-4
- 72.4 Meta-Llama-3-70B-Instruct
- 71.1 Claude 3 Sonnet
- 71.0
Llama 3-70B
- 70.3 Claude 3 Haiku
- 68.4 latxa-70b-v1.1
- 65.9 latxa-70b-v1
- 65.9
Qwen2-72B
- 64.3
Gemma 2 27B
- 63.5 gemma-2-27b-it
- 62.8 GPT-3.5 Turbo
- 62.5 latxa-13b-v1.1
- 62.4
Gemma 2 9B
- 62.1 gemma-2-9b-it
- 61.3
Mixtral 8x7B
- 61.3
Qwen1.5-72B
- 60.6 latxa-13b-v1
- 59.9
Yi-34B
- 59.3 Meta-Llama-3-8B-Instruct
- 58.2
Llama 3-8B
- 57.5
Gemma 7B
- 55.0
Qwen2-7B
- 54.7
Llama 2-70B
- 54.3
Qwen2-7B-Instruct
- 52.9
Qwen1.5-14B
- 52.6
Mistral 7B v0.1
- 52.2 latxa-7b-v1
- 51.6 Yi-9B
- 50.5
Yi 6B
- 50.3 latxa-7b-v1.1
- 49.1
Qwen1.5-7B
- 48.0
Llama 2-13B
- 44.5
Llama 2-7B
- 41.8
Gemma 2B
- 34.6 mGPT-13B
- 34.0 bloom-7b1
- 33.8 xglm-7.5B
- 1 Claude 3 Opus 5-shot 81.5%
- 2 GPT-4 Turbo (Jan 2024) 5-shot 80.7%
- 3 GPT-4 Jun 2023, 5-shot 77.6%
- 4 Meta-Llama-3-70B-Instruct 5-shot 72.4%
- 5 Claude 3 Sonnet 5-shot 71.1%
- 6
Llama 3-70B 5-shot 71%
- 7 Claude 3 Haiku 5-shot 70.3%
- 8 latxa-70b-v1.1 5-shot 68.4%
- 9 latxa-70b-v1 5-shot 65.9%
- 10
Qwen2-72B 5-shot 65.9%
- 11
Gemma 2 27B 5-shot 64.3%
- 12 gemma-2-27b-it 5-shot 63.5%
- 13 GPT-3.5 Turbo Jan 2024, 5-shot 62.8%
- 14 latxa-13b-v1.1 5-shot 62.5%
- 15
Gemma 2 9B 5-shot 62.4%
- 16 gemma-2-9b-it 5-shot 62.1%
- 17
Mixtral 8x7B 5-shot 61.3%
- 18
Qwen1.5-72B 5-shot 61.3%
- 19 latxa-13b-v1 5-shot 60.6%
- 20
Yi-34B 5-shot 59.9%
- 21 Meta-Llama-3-8B-Instruct 5-shot 59.3%
- 22
Llama 3-8B 5-shot 58.2%
- 23
Gemma 7B 5-shot 57.5%
- 24
Qwen2-7B 5-shot 55%
- 25
Llama 2-70B 5-shot 54.7%
- 26
Qwen2-7B-Instruct 5-shot 54.3%
- 27
Qwen1.5-14B 5-shot 52.9%
- 28
Mistral 7B v0.1 5-shot 52.6%
- 29 latxa-7b-v1 5-shot 52.2%
- 30 Yi-9B 5-shot 51.6%
- 31
Yi 6B 5-shot 50.5%
- 32 latxa-7b-v1.1 5-shot 50.3%
- 33
Qwen1.5-7B 5-shot 49.1%
- 34
Llama 2-13B 5-shot 48%
- 35
Llama 2-7B 5-shot 44.5%
- 36
Gemma 2B 5-shot 41.8%
- 37 mGPT-13B 5-shot 34.6%
- 38 bloom-7b1 5-shot 34%
- 39 xglm-7.5B 5-shot 33.8%
Source: The BertaQA paper
- 81.5 Claude 3 Opus
- 80.7 GPT-4 Turbo (Jan 2024)
- 77.6 GPT-4
- 71.1 Claude 3 Sonnet
- 71.0
Llama 3-70B
- 70.3 Claude 3 Haiku
- 65.9
Qwen2-72B
- 64.3
Gemma 2 27B
- 62.8 GPT-3.5 Turbo
- 62.4
Gemma 2 9B
- 61.3
Mixtral 8x7B
- 61.3
Qwen1.5-72B
- 59.9
Yi-34B
- 58.2
Llama 3-8B
- 57.5
Gemma 7B
- 55.0
Qwen2-7B
- 54.7
Llama 2-70B
- 54.3
Qwen2-7B-Instruct
- 52.9
Qwen1.5-14B
- 52.6
Mistral 7B v0.1
- 51.6 Yi-9B
- 50.5
Yi 6B
- 49.1
Qwen1.5-7B
- 48.0
Llama 2-13B
- 44.5
Llama 2-7B
- 41.8
Gemma 2B
- 1 Claude 3 Opus 5-shot 81.5%
- 2 GPT-4 Turbo (Jan 2024) 5-shot 80.7%
- 3 GPT-4 Jun 2023, 5-shot 77.6%
- 4 Claude 3 Sonnet 5-shot 71.1%
- 5
Llama 3-70B 5-shot 71%
- 6 Claude 3 Haiku 5-shot 70.3%
- 7
Qwen2-72B 5-shot 65.9%
- 8
Gemma 2 27B 5-shot 64.3%
- 9 GPT-3.5 Turbo Jan 2024, 5-shot 62.8%
- 10
Gemma 2 9B 5-shot 62.4%
- 11
Mixtral 8x7B 5-shot 61.3%
- 12
Qwen1.5-72B 5-shot 61.3%
- 13
Yi-34B 5-shot 59.9%
- 14
Llama 3-8B 5-shot 58.2%
- 15
Gemma 7B 5-shot 57.5%
- 16
Qwen2-7B 5-shot 55%
- 17
Llama 2-70B 5-shot 54.7%
- 18
Qwen2-7B-Instruct 5-shot 54.3%
- 19
Qwen1.5-14B 5-shot 52.9%
- 20
Mistral 7B v0.1 5-shot 52.6%
- 21 Yi-9B 5-shot 51.6%
- 22
Yi 6B 5-shot 50.5%
- 23
Qwen1.5-7B 5-shot 49.1%
- 24
Llama 2-13B 5-shot 48%
- 25
Llama 2-7B 5-shot 44.5%
- 26
Gemma 2B 5-shot 41.8%
Source: The BertaQA paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3 Opus | 81.5% | 5-shot | Feb 2024 | Official leaderboard |
| 2 | GPT-4 Turbo (Jan 2024) | 80.7% | 5-shot | Jan 2024 | Official leaderboard |
| 3 | GPT-4 Jun 2023 | 77.6% | 5-shot | Jun 2023 | Official leaderboard |
| 4 | Meta-Llama-3-70B-Instruct | 72.4% | 5-shot | Official leaderboard | |
| 5 | Claude 3 Sonnet | 71.1% | 5-shot | Feb 2024 | Official leaderboard |
| 6 | | 71% | 5-shot | Apr 2024 | Official leaderboard |
| 7 | Claude 3 Haiku | 70.3% | 5-shot | Mar 2024 | Official leaderboard |
| 8 | latxa-70b-v1.1 | 68.4% | 5-shot | Official leaderboard | |
| 9 | latxa-70b-v1 | 65.9% | 5-shot | Official leaderboard | |
| 10 | | 65.9% | 5-shot | Jun 2024 | Official leaderboard |
| 11 | | 64.3% | 5-shot | Jun 2024 | Official leaderboard |
| 12 | gemma-2-27b-it | 63.5% | 5-shot | Official leaderboard | |
| 13 | GPT-3.5 Turbo Jan 2024 | 62.8% | 5-shot | Jan 2024 | Official leaderboard |
| 14 | latxa-13b-v1.1 | 62.5% | 5-shot | Official leaderboard | |
| 15 | | 62.4% | 5-shot | Jun 2024 | Official leaderboard |
| 16 | gemma-2-9b-it | 62.1% | 5-shot | Official leaderboard | |
| 17 | | 61.3% | 5-shot | Dec 2023 | Official leaderboard |
| 18 | | 61.3% | 5-shot | Feb 2024 | Official leaderboard |
| 19 | latxa-13b-v1 | 60.6% | 5-shot | Official leaderboard | |
| 20 | | 59.9% | 5-shot | Nov 2023 | Official leaderboard |
| 21 | Meta-Llama-3-8B-Instruct | 59.3% | 5-shot | Official leaderboard | |
| 22 | | 58.2% | 5-shot | Apr 2024 | Official leaderboard |
| 23 | | 57.5% | 5-shot | Feb 2024 | Official leaderboard |
| 24 | | 55% | 5-shot | Jun 2024 | Official leaderboard |
| 25 | | 54.7% | 5-shot | Jul 2023 | Official leaderboard |
| 26 | | 54.3% | 5-shot | Jun 2024 | Official leaderboard |
| 27 | | 52.9% | 5-shot | Feb 2024 | Official leaderboard |
| 28 | | 52.6% | 5-shot | Sep 2023 | Official leaderboard |
| 29 | latxa-7b-v1 | 52.2% | 5-shot | Official leaderboard | |
| 30 | Yi-9B | 51.6% | 5-shot | Mar 2024 | Official leaderboard |
| 31 | | 50.5% | 5-shot | Nov 2023 | Official leaderboard |
| 32 | latxa-7b-v1.1 | 50.3% | 5-shot | Official leaderboard | |
| 33 | | 49.1% | 5-shot | Feb 2024 | Official leaderboard |
| 34 | | 48% | 5-shot | Jul 2023 | Official leaderboard |
| 35 | | 44.5% | 5-shot | Jul 2023 | Official leaderboard |
| 36 | | 41.8% | 5-shot | Feb 2024 | Official leaderboard |
| 37 | mGPT-13B | 34.6% | 5-shot | Official leaderboard | |
| 38 | bloom-7b1 | 34% | 5-shot | Official leaderboard | |
| 39 | xglm-7.5B | 33.8% | 5-shot | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3 Opus | 81.5% | 5-shot | Feb 2024 | Official leaderboard |
| 2 | GPT-4 Turbo (Jan 2024) | 80.7% | 5-shot | Jan 2024 | Official leaderboard |
| 3 | GPT-4 Jun 2023 | 77.6% | 5-shot | Jun 2023 | Official leaderboard |
| 4 | Claude 3 Sonnet | 71.1% | 5-shot | Feb 2024 | Official leaderboard |
| 5 | | 71% | 5-shot | Apr 2024 | Official leaderboard |
| 6 | Claude 3 Haiku | 70.3% | 5-shot | Mar 2024 | Official leaderboard |
| 7 | | 65.9% | 5-shot | Jun 2024 | Official leaderboard |
| 8 | | 64.3% | 5-shot | Jun 2024 | Official leaderboard |
| 9 | GPT-3.5 Turbo Jan 2024 | 62.8% | 5-shot | Jan 2024 | Official leaderboard |
| 10 | | 62.4% | 5-shot | Jun 2024 | Official leaderboard |
| 11 | | 61.3% | 5-shot | Dec 2023 | Official leaderboard |
| 12 | | 61.3% | 5-shot | Feb 2024 | Official leaderboard |
| 13 | | 59.9% | 5-shot | Nov 2023 | Official leaderboard |
| 14 | | 58.2% | 5-shot | Apr 2024 | Official leaderboard |
| 15 | | 57.5% | 5-shot | Feb 2024 | Official leaderboard |
| 16 | | 55% | 5-shot | Jun 2024 | Official leaderboard |
| 17 | | 54.7% | 5-shot | Jul 2023 | Official leaderboard |
| 18 | | 54.3% | 5-shot | Jun 2024 | Official leaderboard |
| 19 | | 52.9% | 5-shot | Feb 2024 | Official leaderboard |
| 20 | | 52.6% | 5-shot | Sep 2023 | Official leaderboard |
| 21 | Yi-9B | 51.6% | 5-shot | Mar 2024 | Official leaderboard |
| 22 | | 50.5% | 5-shot | Nov 2023 | Official leaderboard |
| 23 | | 49.1% | 5-shot | Feb 2024 | Official leaderboard |
| 24 | | 48% | 5-shot | Jul 2023 | Official leaderboard |
| 25 | | 44.5% | 5-shot | Jul 2023 | Official leaderboard |
| 26 | | 41.8% | 5-shot | Feb 2024 | 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 English, Basque and English standard error. GPT-4 Turbo (Jan 2024) leads 1 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
English
26 models we track have a score here.
- 1 GPT-4 Turbo (Jan 2024) ↑1 82%
- 2 Claude 3 Opus ↓1 81.9%
- 3 GPT-4 Jun 2023 80.7%
- 4
Qwen2-72B ↑3 73.2%
- 5 Claude 3 Sonnet ↓1 72.5%
- 6
Llama 3-70B ↓1 72.2%
- 7 Claude 3 Haiku ↓1 71.5%
- 8
Mixtral 8x7B ↑3 70.2%
- 9
Qwen1.5-72B ↑3 69.4%
- 10
Yi-34B ↑3 68.9%
Basque
26 models we track have a score here.
- 1 Claude 3 Opus 81.2%
- 2 GPT-4 Turbo (Jan 2024) 79.4%
- 3 GPT-4 Jun 2023 74.5%
- 4
Llama 3-70B ↑1 69.8%
- 5 Claude 3 Sonnet ↓1 69.8%
- 6 Claude 3 Haiku 69.1%
- 7
Gemma 2 27B ↑1 61.6%
- 8
Gemma 2 9B ↑2 59.4%
- 9
Qwen2-72B ↓2 58.6%
- 10 GPT-3.5 Turbo Jan 2024 ↓1 56.8%
English standard error
20 models we track have a score here.
- 1
Qwen2-72B ↑6 0.64%
- 2
Llama 3-70B ↑3 0.65%
- 3
Mixtral 8x7B ↑8 0.66%
- 4
Qwen1.5-72B ↑8 0.67%
- 5
Yi-34B ↑8 0.67%
- 6
Gemma 2 27B ↑2 0.68%
- 7
Gemma 2 9B ↑3 0.69%
- 8
Llama 3-8B ↑6 0.70%
- 9
Llama 2-70B ↑8 0.70%
- 10
Qwen2-7B ↑6 0.70%
Basque standard error
20 models we track have a score here.
- 1
Llama 3-70B ↑4 0.67%
- 2
Llama 2-7B ↑23 0.70%
- 3
Gemma 2B ↑23 0.70%
- 4
Gemma 2 27B ↑4 0.71%
- 5
Llama 2-13B ↑19 0.71%
- 6
Gemma 2 9B ↑4 0.71%
- 7
Qwen1.5-7B ↑16 0.71%
- 8
Qwen2-72B ↓1 0.71%
- 9
Yi 6B ↑13 0.72%
- 10 Yi-9B ↑11 0.72%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The BertaQA 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 BertaQA 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 Knowledge
- MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects.
Intern-S2-397B 89.8%
- MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy.
Seed1.8 92.3%
- SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy.
Qwen 3.6 Max (Preview) 73.9%
- WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1.
Tencent Hy4 preview 83.9%
- BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly.
Gemma 4 31B IT 74.4%
- HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in June 2024, for Gemma 2 27B. 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 (39 entries).
- Pass Models scored 26 models we track, plus 13 other entries on its leaderboard.
- Warning Newest score The newest came 2 years ago, in June 2024, for Gemma 2 27B.
- Note Frontier models None of the top 10 has a score. The strongest tested is Claude 3 Opus (#173).
- Pass Headroom The best is 81.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 27 rows for models we track and 13 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.