EusExams
Models answer past public service examination questions in Basque. The questions come from several Basque institutions.
- Made by
- Julen Etxaniz and colleagues
- Released
- March 2024
- Kind
- Knowledge in many languages
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- Basque
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
-
Qwen3-8B #154 on the capability index
- Last model added
- April 2025
- Scores refreshed
- November 20th, 2026
BSC-LT/ALIA-40b-instruct-2601 tops the leaderboard with 62.8% Qwen3-8B-Base leads with 51.3% among models released by April 2025
Best score per entry from La Leaderboard’s 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 Qwen3-8B-Base, at 51.3%. This view shows only models we track, each at its best.
The newest model with a score here came out in April 2025, 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
- 62.8 BSC-LT/ALIA-40b-instruct-2601
- 51.3
Qwen3-8B-Base
- 50.8
Gemma 2 9B
- 49.4 google/gemma-2-9b-it
- 46.6
Gemma 3 4B PT
- 46.0
Gemma 3 4B
- 45.8 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8
- 44.9
Llama 3.1-8B
- 44.8 meta-llama/Meta-Llama-3.1-8B-Instruct
- 40.7 BSC-LT/salamandra-7b-instruct
- 39.8
Qwen2.5-7B
- 39.4 IIC/RigoChat-7b-v2
- 39.1 Qwen/Qwen2.5-7B-Instruct
- 38.3 BSC-LT/salamandra-7b
- 38.2 meta-llama/Llama-3.2-3B-Instruct
- 37.6
Llama 3.2 3B
- 35.8
Gemma 2 2B IT
- 34.2
Aya Expanse 8B
- 34.1
Qwen2.5-3B
- 34.0 google/gemma-2-2b
- 34.0 01-ai/Yi-1.5-9B
- 33.5 Qwen/Qwen2.5-3B-Instruct
- 33.0 HiTZ/latxa-7b-v1.2
- 32.8
Yi-1.5-9B-Chat
- 32.8 HuggingFaceTB/SmolLM3-3B-Base
- 32.0 occiglot/occiglot-7b-eu5
- 31.9 occiglot/occiglot-7b-eu5-instruct
- 31.9 occiglot/occiglot-7b-es-en
- 31.7
Mistral 7B v0.3
- 31.5 mistralai/Mistral-7B-Instruct-v0.3
- 31.4 occiglot/occiglot-7b-es-en-instruct
- 31.2 utter-project/EuroLLM-9B
- 31.1 openGPT-X/Teuken-7B-instruct-commercial-v0.4
- 31.0
Qwen2.5-1.5B
- 30.9
Qwen3-4B
- 30.8 HuggingFaceTB/SmolLM3-3B
- 30.5 utter-project/EuroLLM-9B-Instruct
- 30.4 Qwen/Qwen2.5-1.5B-Instruct
- 29.8 allenai/OLMo-2-1124-7B
- 29.8 meta-llama/Llama-3.2-1B-Instruct
- 29.2
Gemma 3 1B
- 29.1
SmolLM2-1.7B
- 29.1 bertin-project/Gromenauer-7B-Instruct
- 29.1 allenai/OLMo-2-1124-7B-Instruct
- 29.1 openGPT-X/Teuken-7B-instruct-research-v0.4
- 28.1
SmolLM2 1.7B Instruct
- 28.0 bertin-project/Gromenauer-7B
- 27.8 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0
- 27.5
Llama 3.2 1B
- 26.2 projecte-aina/FLOR-1.3B
- 26.0 proxectonos/Carballo-bloom-1.3B
- 26.0 utter-project/EuroLLM-1.7B
- 26.0
Gemma 3 1B PT
- 25.8 utter-project/EuroLLM-1.7B-Instruct
- 25.8 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit
- 25.6 LenguajeNaturalAI/leniachat-gemma-2b-v0
- 25.6
Phi-1.5
- 25.6 projecte-aina/aguila-7b
- 25.6
Qwen3-8B
- 25.5 projecte-aina/FLOR-6.3B-Instructed
- 25.5 gplsi/Aitana-6.3B
- 25.3 bertin-project/bertin-gpt-j-6B
- 25.3 projecte-aina/FLOR-1.3B-Instructed
- 25.1 projecte-aina/FLOR-6.3B
- 25.0
DeepSeek-R1-Distill-Qwen-1.5B
- 24.9 BSC-LT/salamandra-2b-instruct
- 24.8
DeepSeek-R1-Distill-Qwen-7B
- 24.6
Qwen3-1.7B
- 23.7 BSC-LT/salamandra-2b
- 1 BSC-LT/ALIA-40b-instruct-2601 4-shot 62.8%
- 2
Qwen3-8B-Base 51.3%
- 3
Gemma 2 9B 4-shot 50.8%
- 4 google/gemma-2-9b-it 4-shot 49.4%
- 5
Gemma 3 4B PT 46.6%
- 6
Gemma 3 4B 46%
- 7 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8 45.8%
- 8
Llama 3.1-8B 4-shot 44.9%
- 9 meta-llama/Meta-Llama-3.1-8B-Instruct 4-shot 44.8%
- 10 BSC-LT/salamandra-7b-instruct 4-shot 40.7%
- 11
Qwen2.5-7B 4-shot 39.8%
- 12 IIC/RigoChat-7b-v2 4-shot 39.4%
- 13 Qwen/Qwen2.5-7B-Instruct 4-shot 39.1%
- 14 BSC-LT/salamandra-7b 4-shot 38.3%
- 15 meta-llama/Llama-3.2-3B-Instruct 38.2%
- 16
Llama 3.2 3B 37.6%
- 17
Gemma 2 2B IT 4-shot 35.8%
- 18
Aya Expanse 8B 4-shot 34.2%
- 19
Qwen2.5-3B 34.1%
- 20 google/gemma-2-2b 4-shot 34%
- 21 01-ai/Yi-1.5-9B 4-shot 34%
- 22 Qwen/Qwen2.5-3B-Instruct 33.5%
- 23 HiTZ/latxa-7b-v1.2 4-shot 33%
- 24
Yi-1.5-9B-Chat 4-shot 32.8%
- 25 HuggingFaceTB/SmolLM3-3B-Base 32.8%
- 26 occiglot/occiglot-7b-eu5 4-shot 32%
- 27 occiglot/occiglot-7b-eu5-instruct 4-shot 31.9%
- 28 occiglot/occiglot-7b-es-en 4-shot 31.9%
- 29
Mistral 7B v0.3 4-shot 31.7%
- 30 mistralai/Mistral-7B-Instruct-v0.3 4-shot 31.5%
- 31 occiglot/occiglot-7b-es-en-instruct 4-shot 31.4%
- 32 utter-project/EuroLLM-9B 4-shot 31.2%
- 33 openGPT-X/Teuken-7B-instruct-commercial-v0.4 31.1%
- 34
Qwen2.5-1.5B 4-shot 31%
- 35
Qwen3-4B 30.9%
- 36 HuggingFaceTB/SmolLM3-3B 30.8%
- 37 utter-project/EuroLLM-9B-Instruct 4-shot 30.5%
- 38 Qwen/Qwen2.5-1.5B-Instruct 4-shot 30.4%
- 39 allenai/OLMo-2-1124-7B 29.8%
- 40 meta-llama/Llama-3.2-1B-Instruct 4-shot 29.8%
- 41
Gemma 3 1B 29.2%
- 42
SmolLM2-1.7B 29.1%
- 43 bertin-project/Gromenauer-7B-Instruct 4-shot 29.1%
- 44 allenai/OLMo-2-1124-7B-Instruct 29.1%
- 45 openGPT-X/Teuken-7B-instruct-research-v0.4 29.1%
- 46
SmolLM2 1.7B Instruct 28.1%
- 47 bertin-project/Gromenauer-7B 4-shot 28%
- 48 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0 4-shot 27.8%
- 49
Llama 3.2 1B 4-shot 27.5%
- 50 projecte-aina/FLOR-1.3B 4-shot 26.2%
- 51 proxectonos/Carballo-bloom-1.3B 4-shot 26%
- 52 utter-project/EuroLLM-1.7B 4-shot 26%
- 53
Gemma 3 1B PT 26%
- 54 utter-project/EuroLLM-1.7B-Instruct 4-shot 25.8%
- 55 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit 4-shot 25.8%
- 56 LenguajeNaturalAI/leniachat-gemma-2b-v0 4-shot 25.6%
- 57
Phi-1.5 4-shot 25.6%
- 58 projecte-aina/aguila-7b 4-shot 25.6%
- 59
Qwen3-8B 25.6%
- 60 projecte-aina/FLOR-6.3B-Instructed 4-shot 25.5%
- 61 gplsi/Aitana-6.3B 4-shot 25.5%
- 62 bertin-project/bertin-gpt-j-6B 4-shot 25.3%
- 63 projecte-aina/FLOR-1.3B-Instructed 4-shot 25.3%
- 64 projecte-aina/FLOR-6.3B 4-shot 25.1%
- 65
DeepSeek-R1-Distill-Qwen-1.5B 4-shot 25%
- 66 BSC-LT/salamandra-2b-instruct 4-shot 24.9%
- 67
DeepSeek-R1-Distill-Qwen-7B 4-shot 24.8%
- 68
Qwen3-1.7B 24.6%
- 69 BSC-LT/salamandra-2b 4-shot 23.7%
Source: La Leaderboard’s leaderboard
- 51.3
Qwen3-8B-Base
- 50.8
Gemma 2 9B
- 46.6
Gemma 3 4B PT
- 46.0
Gemma 3 4B
- 44.9
Llama 3.1-8B
- 39.8
Qwen2.5-7B
- 37.6
Llama 3.2 3B
- 35.8
Gemma 2 2B IT
- 34.2
Aya Expanse 8B
- 34.1
Qwen2.5-3B
- 32.8
Yi-1.5-9B-Chat
- 31.7
Mistral 7B v0.3
- 31.0
Qwen2.5-1.5B
- 30.9
Qwen3-4B
- 29.2
Gemma 3 1B
- 29.1
SmolLM2-1.7B
- 28.1
SmolLM2 1.7B Instruct
- 27.5
Llama 3.2 1B
- 26.0
Gemma 3 1B PT
- 25.6
Phi-1.5
- 25.6
Qwen3-8B
- 25.0
DeepSeek-R1-Distill-Qwen-1.5B
- 24.8
DeepSeek-R1-Distill-Qwen-7B
- 24.6
Qwen3-1.7B
- 1
Qwen3-8B-Base 51.3%
- 2
Gemma 2 9B 4-shot 50.8%
- 3
Gemma 3 4B PT 46.6%
- 4
Gemma 3 4B 46%
- 5
Llama 3.1-8B 4-shot 44.9%
- 6
Qwen2.5-7B 4-shot 39.8%
- 7
Llama 3.2 3B 37.6%
- 8
Gemma 2 2B IT 4-shot 35.8%
- 9
Aya Expanse 8B 4-shot 34.2%
- 10
Qwen2.5-3B 34.1%
- 11
Yi-1.5-9B-Chat 4-shot 32.8%
- 12
Mistral 7B v0.3 4-shot 31.7%
- 13
Qwen2.5-1.5B 4-shot 31%
- 14
Qwen3-4B 30.9%
- 15
Gemma 3 1B 29.2%
- 16
SmolLM2-1.7B 29.1%
- 17
SmolLM2 1.7B Instruct 28.1%
- 18
Llama 3.2 1B 4-shot 27.5%
- 19
Gemma 3 1B PT 26%
- 20
Phi-1.5 4-shot 25.6%
- 21
Qwen3-8B 25.6%
- 22
DeepSeek-R1-Distill-Qwen-1.5B 4-shot 25%
- 23
DeepSeek-R1-Distill-Qwen-7B 4-shot 24.8%
- 24
Qwen3-1.7B 24.6%
Source: La Leaderboard’s leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | BSC-LT/ALIA-40b-instruct-2601 | 62.8% | 4-shot | Official leaderboard | |
| 2 | | 51.3% | Apr 2025 | Official leaderboard | |
| 3 | | 50.8% | 4-shot | Jun 2024 | Official leaderboard |
| 4 | google/gemma-2-9b-it | 49.4% | 4-shot | Official leaderboard | |
| 5 | | 46.6% | Mar 2025 | Official leaderboard | |
| 6 | | 46% | Mar 2025 | Official leaderboard | |
| 7 | Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8 | 45.8% | Official leaderboard | ||
| 8 | | 44.9% | 4-shot | Jul 2024 | Official leaderboard |
| 9 | meta-llama/Meta-Llama-3.1-8B-Instruct | 44.8% | 4-shot | Official leaderboard | |
| 10 | BSC-LT/salamandra-7b-instruct | 40.7% | 4-shot | Official leaderboard | |
| 11 | | 39.8% | 4-shot | Sep 2024 | Official leaderboard |
| 12 | IIC/RigoChat-7b-v2 | 39.4% | 4-shot | Official leaderboard | |
| 13 | Qwen/Qwen2.5-7B-Instruct | 39.1% | 4-shot | Official leaderboard | |
| 14 | BSC-LT/salamandra-7b | 38.3% | 4-shot | Official leaderboard | |
| 15 | meta-llama/Llama-3.2-3B-Instruct | 38.2% | Official leaderboard | ||
| 16 | | 37.6% | Sep 2024 | Official leaderboard | |
| 17 | | 35.8% | 4-shot | Jul 2024 | Official leaderboard |
| 18 | | 34.2% | 4-shot | Oct 2024 | Official leaderboard |
| 19 | | 34.1% | Sep 2024 | Official leaderboard | |
| 20 | google/gemma-2-2b | 34% | 4-shot | Official leaderboard | |
| 21 | 01-ai/Yi-1.5-9B | 34% | 4-shot | Official leaderboard | |
| 22 | Qwen/Qwen2.5-3B-Instruct | 33.5% | Official leaderboard | ||
| 23 | HiTZ/latxa-7b-v1.2 | 33% | 4-shot | Official leaderboard | |
| 24 | | 32.8% | 4-shot | May 2024 | Official leaderboard |
| 25 | HuggingFaceTB/SmolLM3-3B-Base | 32.8% | Official leaderboard | ||
| 26 | occiglot/occiglot-7b-eu5 | 32% | 4-shot | Official leaderboard | |
| 27 | occiglot/occiglot-7b-eu5-instruct | 31.9% | 4-shot | Official leaderboard | |
| 28 | occiglot/occiglot-7b-es-en | 31.9% | 4-shot | Official leaderboard | |
| 29 | | 31.7% | 4-shot | Sep 2023 | Official leaderboard |
| 30 | mistralai/Mistral-7B-Instruct-v0.3 | 31.5% | 4-shot | Official leaderboard | |
| 31 | occiglot/occiglot-7b-es-en-instruct | 31.4% | 4-shot | Official leaderboard | |
| 32 | utter-project/EuroLLM-9B | 31.2% | 4-shot | Official leaderboard | |
| 33 | openGPT-X/Teuken-7B-instruct-commercial-v0.4 | 31.1% | Official leaderboard | ||
| 34 | | 31% | 4-shot | Sep 2024 | Official leaderboard |
| 35 | | 30.9% | Apr 2025 | Official leaderboard | |
| 36 | HuggingFaceTB/SmolLM3-3B | 30.8% | Official leaderboard | ||
| 37 | utter-project/EuroLLM-9B-Instruct | 30.5% | 4-shot | Official leaderboard | |
| 38 | Qwen/Qwen2.5-1.5B-Instruct | 30.4% | 4-shot | Official leaderboard | |
| 39 | allenai/OLMo-2-1124-7B | 29.8% | Official leaderboard | ||
| 40 | meta-llama/Llama-3.2-1B-Instruct | 29.8% | 4-shot | Official leaderboard | |
| 41 | | 29.2% | Mar 2025 | Official leaderboard | |
| 42 | | 29.1% | Oct 2024 | Official leaderboard | |
| 43 | bertin-project/Gromenauer-7B-Instruct | 29.1% | 4-shot | Official leaderboard | |
| 44 | allenai/OLMo-2-1124-7B-Instruct | 29.1% | Official leaderboard | ||
| 45 | openGPT-X/Teuken-7B-instruct-research-v0.4 | 29.1% | Official leaderboard | ||
| 46 | | 28.1% | Oct 2024 | Official leaderboard | |
| 47 | bertin-project/Gromenauer-7B | 28% | 4-shot | Official leaderboard | |
| 48 | LenguajeNaturalAI/leniachat-qwen2-1.5B-v0 | 27.8% | 4-shot | Official leaderboard | |
| 49 | | 27.5% | 4-shot | Sep 2024 | Official leaderboard |
| 50 | projecte-aina/FLOR-1.3B | 26.2% | 4-shot | Official leaderboard | |
| 51 | proxectonos/Carballo-bloom-1.3B | 26% | 4-shot | Official leaderboard | |
| 52 | utter-project/EuroLLM-1.7B | 26% | 4-shot | Official leaderboard | |
| 53 | | 26% | Mar 2025 | Official leaderboard | |
| 54 | utter-project/EuroLLM-1.7B-Instruct | 25.8% | 4-shot | Official leaderboard | |
| 55 | unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit | 25.8% | 4-shot | Official leaderboard | |
| 56 | LenguajeNaturalAI/leniachat-gemma-2b-v0 | 25.6% | 4-shot | Official leaderboard | |
| 57 | | 25.6% | 4-shot | Sep 2023 | Official leaderboard |
| 58 | projecte-aina/aguila-7b | 25.6% | 4-shot | Official leaderboard | |
| 59 | | 25.6% | Apr 2025 | Official leaderboard | |
| 60 | projecte-aina/FLOR-6.3B-Instructed | 25.5% | 4-shot | Official leaderboard | |
| 61 | gplsi/Aitana-6.3B | 25.5% | 4-shot | Official leaderboard | |
| 62 | bertin-project/bertin-gpt-j-6B | 25.3% | 4-shot | Official leaderboard | |
| 63 | projecte-aina/FLOR-1.3B-Instructed | 25.3% | 4-shot | Official leaderboard | |
| 64 | projecte-aina/FLOR-6.3B | 25.1% | 4-shot | Official leaderboard | |
| 65 | | 25% | 4-shot | Jan 2025 | Official leaderboard |
| 66 | BSC-LT/salamandra-2b-instruct | 24.9% | 4-shot | Official leaderboard | |
| 67 | | 24.8% | 4-shot | Jan 2025 | Official leaderboard |
| 68 | | 24.6% | Apr 2025 | Official leaderboard | |
| 69 | BSC-LT/salamandra-2b | 23.7% | 4-shot | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 51.3% | Apr 2025 | Official leaderboard | |
| 2 | | 50.8% | 4-shot | Jun 2024 | Official leaderboard |
| 3 | | 46.6% | Mar 2025 | Official leaderboard | |
| 4 | | 46% | Mar 2025 | Official leaderboard | |
| 5 | | 44.9% | 4-shot | Jul 2024 | Official leaderboard |
| 6 | | 39.8% | 4-shot | Sep 2024 | Official leaderboard |
| 7 | | 37.6% | Sep 2024 | Official leaderboard | |
| 8 | | 35.8% | 4-shot | Jul 2024 | Official leaderboard |
| 9 | | 34.2% | 4-shot | Oct 2024 | Official leaderboard |
| 10 | | 34.1% | Sep 2024 | Official leaderboard | |
| 11 | | 32.8% | 4-shot | May 2024 | Official leaderboard |
| 12 | | 31.7% | 4-shot | Sep 2023 | Official leaderboard |
| 13 | | 31% | 4-shot | Sep 2024 | Official leaderboard |
| 14 | | 30.9% | Apr 2025 | Official leaderboard | |
| 15 | | 29.2% | Mar 2025 | Official leaderboard | |
| 16 | | 29.1% | Oct 2024 | Official leaderboard | |
| 17 | | 28.1% | Oct 2024 | Official leaderboard | |
| 18 | | 27.5% | 4-shot | Sep 2024 | Official leaderboard |
| 19 | | 26% | Mar 2025 | Official leaderboard | |
| 20 | | 25.6% | 4-shot | Sep 2023 | Official leaderboard |
| 21 | | 25.6% | Apr 2025 | Official leaderboard | |
| 22 | | 25% | 4-shot | Jan 2025 | Official leaderboard |
| 23 | | 24.8% | 4-shot | Jan 2025 | Official leaderboard |
| 24 | | 24.6% | Apr 2025 | Official leaderboard |
Rankings by part
We read 29 parts from La Leaderboard’s leaderboard, each ranking models on one part of the test, such as ejadministrari, ejlaguntza and ejlaguntzaile. Qwen3-8B-Base leads 25 of them, and Gemma 3 4B PT leads the other 4. Arrows show how far a model moved from its overall rank among the models we track.
ejadministrari
12 models we track have a score here.
- 1
Qwen3-8B-Base 53.4%
- 2
Gemma 3 4B PT ↑1 48.7%
- 3
Gemma 3 4B ↑1 48.4%
- 4
Llama 3.2 3B ↑3 41.1%
- 5
Qwen2.5-3B ↑5 36.2%
- 6
Qwen3-4B ↑8 32.2%
- 7
Gemma 3 1B ↑8 31.3%
- 8
SmolLM2 1.7B Instruct ↑9 30.6%
- 9
SmolLM2-1.7B ↑7 29.9%
- 10
Gemma 3 1B PT ↑9 26%
ejlaguntza
12 models we track have a score here.
- 1
Qwen3-8B-Base 55.6%
- 2
Gemma 3 4B PT ↑1 52.6%
- 3
Gemma 3 4B ↑1 50.2%
- 4
Llama 3.2 3B ↑3 42.2%
- 5
Gemma 3 1B ↑10 36.7%
- 6
Qwen2.5-3B ↑4 34.9%
- 7
SmolLM2 1.7B Instruct ↑10 30.7%
- 8
Qwen3-4B ↑6 29.1%
- 9
SmolLM2-1.7B ↑7 28.7%
- 10
Gemma 3 1B PT ↑9 25.9%
ejlaguntzaile
12 models we track have a score here.
- 1
Gemma 3 4B PT ↑2 54.1%
- 2
Qwen3-8B-Base ↓1 52.1%
- 3
Gemma 3 4B ↑1 49.9%
- 4
Llama 3.2 3B ↑3 42.5%
- 5
Qwen2.5-3B ↑5 36.9%
- 6
SmolLM2-1.7B ↑10 33.4%
- 7
Gemma 3 1B ↑8 30.3%
- 8
Qwen3-4B ↑6 29.8%
- 9
SmolLM2 1.7B Instruct ↑8 28.8%
- 10
Gemma 3 1B PT ↑9 27.3%
ejteknikari
12 models we track have a score here.
- 1
Qwen3-8B-Base 55.3%
- 2
Gemma 3 4B PT ↑1 54.6%
- 3
Gemma 3 4B ↑1 52%
- 4
Llama 3.2 3B ↑3 43.1%
- 5
Qwen2.5-3B ↑5 37.1%
- 6
Qwen3-4B ↑8 33.2%
- 7
Gemma 3 1B ↑8 32.8%
- 8
SmolLM2 1.7B Instruct ↑9 30.7%
- 9
SmolLM2-1.7B ↑7 28.2%
- 10
Qwen3-8B ↑11 26.8%
opebilbaoeu
12 models we track have a score here.
- 1
Gemma 3 4B PT ↑2 49.4%
- 2
Qwen3-8B-Base ↓1 48.6%
- 3
Gemma 3 4B ↑1 42.2%
- 4
Llama 3.2 3B ↑3 36.7%
- 5
Qwen2.5-3B ↑5 31.4%
- 6
Gemma 3 1B ↑9 30.6%
- 7
Qwen3-4B ↑7 30.5%
- 8
SmolLM2 1.7B Instruct ↑9 28.4%
- 9
SmolLM2-1.7B ↑7 28.3%
- 10
Gemma 3 1B PT ↑9 28.1%
opeehuadmineu
12 models we track have a score here.
- 1
Qwen3-8B-Base 56.7%
- 2
Gemma 3 4B ↑2 55.1%
- 3
Gemma 3 4B PT 52.9%
- 4
Llama 3.2 3B ↑3 40.9%
- 5
Qwen2.5-3B ↑5 39.1%
- 6
Qwen3-4B ↑8 32.3%
- 7
Gemma 3 1B ↑8 31.1%
- 8
SmolLM2-1.7B ↑8 30.1%
- 9
SmolLM2 1.7B Instruct ↑8 29.7%
- 10
Qwen3-8B ↑11 26.7%
opeehuauxeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 59.1%
- 2
Gemma 3 4B PT ↑1 54.9%
- 3
Gemma 3 4B ↑1 54%
- 4
Llama 3.2 3B ↑3 44.4%
- 5
Qwen2.5-3B ↑5 38%
- 6
Qwen3-4B ↑8 32.9%
- 7
Gemma 3 1B ↑8 32.2%
- 8
SmolLM2 1.7B Instruct ↑9 30.7%
- 9
SmolLM2-1.7B ↑7 30%
- 10
Qwen3-8B ↑11 26.2%
opeehubiblioeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 51.6%
- 2
Gemma 3 4B ↑2 50.1%
- 3
Gemma 3 4B PT 50.1%
- 4
Llama 3.2 3B ↑3 38.9%
- 5
Qwen2.5-3B ↑5 36.9%
- 6
Qwen3-4B ↑8 31.2%
- 7
Gemma 3 1B ↑8 29.9%
- 8
SmolLM2-1.7B ↑8 27.9%
- 9
SmolLM2 1.7B Instruct ↑8 25.2%
- 10
Gemma 3 1B PT ↑9 25.2%
opeehuderechoeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 47%
- 2
Gemma 3 4B ↑2 45%
- 3
Gemma 3 4B PT 43.7%
- 4
Llama 3.2 3B ↑3 35.3%
- 5
Qwen2.5-3B ↑5 33.1%
- 6
SmolLM2 1.7B Instruct ↑11 28.6%
- 7
Qwen3-4B ↑7 27.4%
- 8
SmolLM2-1.7B ↑8 27.1%
- 9
Gemma 3 1B ↑6 25.3%
- 10
Gemma 3 1B PT ↑9 23.3%
opeehueconomicaseu
12 models we track have a score here.
- 1
Qwen3-8B-Base 53.3%
- 2
Gemma 3 4B ↑2 48.7%
- 3
Gemma 3 4B PT 48.4%
- 4
Qwen2.5-3B ↑6 36.5%
- 5
Llama 3.2 3B ↑2 35.6%
- 6
Gemma 3 1B ↑9 33.6%
- 7
SmolLM2-1.7B ↑9 27.6%
- 8
SmolLM2 1.7B Instruct ↑9 27.1%
- 9
Qwen3-1.7B ↑15 25.6%
- 10
Qwen3-8B ↑11 25.4%
opeehuempresarialeseu
12 models we track have a score here.
- 1
Qwen3-8B-Base 52.1%
- 2
Gemma 3 4B ↑2 51.4%
- 3
Gemma 3 4B PT 48.9%
- 4
Llama 3.2 3B ↑3 40%
- 5
Qwen2.5-3B ↑5 36.8%
- 6
Qwen3-4B ↑8 33.2%
- 7
Gemma 3 1B ↑8 30.4%
- 8
SmolLM2-1.7B ↑8 29.6%
- 9
Qwen3-8B ↑12 28.2%
- 10
SmolLM2 1.7B Instruct ↑7 27.5%
opeehusubalternoeu
12 models we track have a score here.
- 1
Gemma 3 4B PT ↑2 59.8%
- 2
Qwen3-8B-Base ↓1 59.3%
- 3
Gemma 3 4B ↑1 54%
- 4
Llama 3.2 3B ↑3 43.8%
- 5
Qwen2.5-3B ↑5 39.8%
- 6
Qwen3-4B ↑8 35%
- 7
SmolLM2-1.7B ↑9 32%
- 8
Gemma 3 1B ↑7 31.3%
- 9
Qwen3-8B ↑12 28.3%
- 10
SmolLM2 1.7B Instruct ↑7 28%
opeehutecnicoeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 51.6%
- 2
Gemma 3 4B ↑2 48.6%
- 3
Gemma 3 4B PT 46.5%
- 4
Llama 3.2 3B ↑3 38.6%
- 5
Qwen2.5-3B ↑5 35.3%
- 6
Qwen3-4B ↑8 30.9%
- 7
SmolLM2 1.7B Instruct ↑10 29.8%
- 8
Gemma 3 1B ↑7 27.8%
- 9
SmolLM2-1.7B ↑7 27.6%
- 10
Gemma 3 1B PT ↑9 26.2%
opeehuteknikarib
12 models we track have a score here.
- 1
Qwen3-8B-Base 53.1%
- 2
Gemma 3 4B ↑2 51.3%
- 3
Gemma 3 4B PT 49.4%
- 4
Llama 3.2 3B ↑3 39.2%
- 5
Qwen2.5-3B ↑5 38.4%
- 6
Gemma 3 1B ↑9 31.2%
- 7
SmolLM2-1.7B ↑9 30.9%
- 8
SmolLM2 1.7B Instruct ↑9 29.7%
- 9
Qwen3-4B ↑5 28.2%
- 10
Qwen3-8B ↑11 25.4%
opegasteizkoudala
12 models we track have a score here.
- 1
Qwen3-8B-Base 51.9%
- 2
Gemma 3 4B ↑2 45.6%
- 3
Gemma 3 4B PT 41.7%
- 4
Llama 3.2 3B ↑3 38.1%
- 5
Qwen3-8B ↑16 31.4%
- 6
Qwen2.5-3B ↑4 31.1%
- 7
Qwen3-4B ↑7 29.2%
- 8
Gemma 3 1B ↑7 28.3%
- 9
Qwen3-1.7B ↑15 27.8%
- 10
SmolLM2-1.7B ↑6 25.8%
opeosakiadmineu
12 models we track have a score here.
- 1
Qwen3-8B-Base 46.2%
- 2
Gemma 3 4B ↑2 39.4%
- 3
Gemma 3 4B PT 36%
- 4
Llama 3.2 3B ↑3 32.6%
- 5
Qwen3-4B ↑9 31.8%
- 6
Qwen3-8B ↑15 30.9%
- 7
Qwen3-1.7B ↑17 30.1%
- 8
SmolLM2-1.7B ↑8 29.7%
- 9
Gemma 3 1B ↑6 29.2%
- 10
Qwen2.5-3B 27.5%
opeosakiauxenfeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 47.3%
- 2
Gemma 3 4B PT ↑1 33.5%
- 3
Qwen3-4B ↑11 32.3%
- 4
Gemma 3 4B 31.7%
- 5
Qwen3-8B ↑16 31.1%
- 6
Qwen3-1.7B ↑18 30.5%
- 7
Llama 3.2 3B 29.9%
- 8
SmolLM2-1.7B ↑8 28.7%
- 9
Gemma 3 1B ↑6 28.7%
- 10
Gemma 3 1B PT ↑9 27.5%
opeosakiauxeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 52.1%
- 2
Gemma 3 4B ↑2 34.5%
- 3
Qwen3-4B ↑11 34.5%
- 4
Qwen3-8B ↑17 32.7%
- 5
Gemma 3 4B PT ↓2 31.5%
- 6
Qwen3-1.7B ↑18 30.9%
- 7
Gemma 3 1B PT ↑12 30.3%
- 8
Llama 3.2 3B ↓1 28.5%
- 9
Gemma 3 1B ↑6 27.9%
- 10
SmolLM2-1.7B ↑6 26.7%
opeosakiceladoreu
12 models we track have a score here.
- 1
Qwen3-8B-Base 47.9%
- 2
Gemma 3 4B ↑2 35.6%
- 3
Qwen3-8B ↑18 31.9%
- 4
Qwen3-1.7B ↑20 30.7%
- 5
Qwen3-4B ↑9 30.7%
- 6
Gemma 3 4B PT ↓3 30.1%
- 7
Qwen2.5-3B ↑3 28.2%
- 8
Llama 3.2 3B ↓1 28.2%
- 9
Gemma 3 1B ↑6 26.4%
- 10
SmolLM2-1.7B ↑6 25.2%
opeosakienfeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 45.6%
- 2
Gemma 3 4B ↑2 38.9%
- 3
Gemma 3 4B PT 35.4%
- 4
Llama 3.2 3B ↑3 31.9%
- 5
Qwen3-8B ↑16 30.3%
- 6
Qwen3-1.7B ↑18 30.3%
- 7
Qwen3-4B ↑7 30%
- 8
SmolLM2-1.7B ↑8 28.4%
- 9
Qwen2.5-3B ↑1 27.9%
- 10
Gemma 3 1B ↑5 27.1%
opeosakioperarioeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 50.4%
- 2
Gemma 3 4B ↑2 36.6%
- 3
Qwen3-4B ↑11 35.1%
- 4
Gemma 3 4B PT ↓1 33.6%
- 5
Qwen3-8B ↑16 30.5%
- 6
Llama 3.2 3B ↑1 29.8%
- 7
Qwen3-1.7B ↑17 29.8%
- 8
Gemma 3 1B PT ↑11 28.2%
- 9
Qwen2.5-3B ↑1 26%
- 10
Gemma 3 1B ↑5 24.4%
opeosakitecnicoeu
12 models we track have a score here.
- 1
Qwen3-8B-Base 45.2%
- 2
Gemma 3 4B ↑2 37.4%
- 3
Gemma 3 4B PT 35.8%
- 4
Qwen3-4B ↑10 33%
- 5
Llama 3.2 3B ↑2 32.7%
- 6
SmolLM2-1.7B ↑10 32.7%
- 7
Qwen3-8B ↑14 31.2%
- 8
Qwen3-1.7B ↑16 30.8%
- 9
Gemma 3 1B PT ↑10 29.9%
- 10
Qwen2.5-3B 28%
opeosakivarioseu
12 models we track have a score here.
- 1
Qwen3-8B-Base 46.6%
- 2
Gemma 3 4B ↑2 36.5%
- 3
Gemma 3 4B PT 33.5%
- 4
Qwen3-8B ↑17 32%
- 5
Qwen3-1.7B ↑19 31.6%
- 6
Qwen3-4B ↑8 30.8%
- 7
Gemma 3 1B PT ↑12 30.5%
- 8
Llama 3.2 3B ↓1 30.1%
- 9
Qwen2.5-3B ↑1 29.3%
- 10
SmolLM2-1.7B ↑6 28.2%
osakidetza1e
12 models we track have a score here.
- 1
Qwen3-8B-Base 37.6%
- 2
Gemma 3 4B PT ↑1 35.1%
- 3
Gemma 3 4B ↑1 32.9%
- 4
Llama 3.2 3B ↑3 31.5%
- 5
SmolLM2 1.7B Instruct ↑12 30.1%
- 6
Qwen3-4B ↑8 28.1%
- 7
Qwen2.5-3B ↑3 27.5%
- 8
Gemma 3 1B ↑7 26.9%
- 9
Gemma 3 1B PT ↑10 26.9%
- 10
SmolLM2-1.7B ↑6 26.7%
osakidetza2e
12 models we track have a score here.
- 1
Gemma 3 4B PT ↑2 38.6%
- 2
Qwen3-8B-Base ↓1 38.2%
- 3
Gemma 3 4B ↑1 35.4%
- 4
Llama 3.2 3B ↑3 33.2%
- 5
Qwen2.5-3B ↑5 31.1%
- 6
SmolLM2 1.7B Instruct ↑11 27.9%
- 7
SmolLM2-1.7B ↑9 27.8%
- 8
Gemma 3 1B ↑7 27.1%
- 9
Qwen3-4B ↑5 27.1%
- 10
Gemma 3 1B PT ↑9 25.5%
osakidetza3e
12 models we track have a score here.
- 1
Qwen3-8B-Base 49.9%
- 2
Gemma 3 4B PT ↑1 48.3%
- 3
Gemma 3 4B ↑1 46.1%
- 4
Llama 3.2 3B ↑3 36.1%
- 5
Qwen2.5-3B ↑5 35.9%
- 6
SmolLM2 1.7B Instruct ↑11 32.8%
- 7
SmolLM2-1.7B ↑9 29.9%
- 8
Gemma 3 1B ↑7 29.5%
- 9
Qwen3-4B ↑5 26.8%
- 10
Qwen3-8B ↑11 25.3%
osakidetza5e
12 models we track have a score here.
- 1
Qwen3-8B-Base 55.2%
- 2
Gemma 3 4B PT ↑1 49.8%
- 3
Gemma 3 4B ↑1 46.8%
- 4
Llama 3.2 3B ↑3 39.8%
- 5
Qwen2.5-3B ↑5 37.4%
- 6
Qwen3-4B ↑8 34.3%
- 7
Gemma 3 1B ↑8 31.7%
- 8
SmolLM2 1.7B Instruct ↑9 30.9%
- 9
SmolLM2-1.7B ↑7 30%
- 10
Gemma 3 1B PT ↑9 28.2%
osakidetza6e
12 models we track have a score here.
- 1
Qwen3-8B-Base 59.2%
- 2
Gemma 3 4B PT ↑1 51.4%
- 3
Gemma 3 4B ↑1 50.1%
- 4
Llama 3.2 3B ↑3 40.1%
- 5
Qwen2.5-3B ↑5 36%
- 6
SmolLM2-1.7B ↑10 32.8%
- 7
Qwen3-4B ↑7 32.8%
- 8
Gemma 3 1B PT ↑11 29.2%
- 9
SmolLM2 1.7B Instruct ↑8 27.5%
- 10
Gemma 3 1B ↑5 26.9%
osakidetza7e
12 models we track have a score here.
- 1
Qwen3-8B-Base 52.1%
- 2
Gemma 3 4B PT ↑1 46.1%
- 3
Gemma 3 4B ↑1 45.8%
- 4
Llama 3.2 3B ↑3 36.2%
- 5
Qwen2.5-3B ↑5 32.9%
- 6
Qwen3-4B ↑8 32.3%
- 7
SmolLM2 1.7B Instruct ↑10 29.3%
- 8
SmolLM2-1.7B ↑8 28.6%
- 9
Gemma 3 1B ↑6 27.6%
- 10
Gemma 3 1B PT ↑9 25.2%
4 models have held number one on La Leaderboard’s leaderboard since its release in March 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: La Leaderboard’s leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: La Leaderboard’s 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 19 months ago, in April 2025, for Qwen3-1.7B. 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 (69 entries).
- Pass Models scored 24 models we track, plus 45 other entries on its leaderboard.
- Warning Newest score The newest came 19 months ago, in April 2025, for Qwen3-1.7B.
- Note Frontier models None of the top 10 has a score. The strongest tested is Qwen3-8B (#154).
- Pass Headroom The best is 51.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on La Leaderboard: 24 rows for models we track and 45 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.