Benchmark Almanac

EusTrivia

BSC-LT/ALIA-40b-instruct-2601 66.2% tops its leaderboard Gemma 2 9B 51% leads the models we track
Not in model cards

Models answer trivia questions in Basque. The set combines elementary-school questions with harder general knowledge questions.

Released
March 2024
Kind
Knowledge in many languages
Level
School
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
Basque
In and out
Text → Text
Interaction
Single turn
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
69
Entries on La Leaderboard’s leaderboard, 24 models we track

BSC-LT/ALIA-40b-instruct-2601 tops the leaderboard with 66.2% Gemma 2 9B leads with 51% 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 Gemma 2 9B, at 51%. 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:

69 entries. Scroll sideways for the rest.

  1. 66.2 BSC-LT/ALIA-40b-instruct-2601
  2. 51.4 google/gemma-2-9b-it
  3. 51.0 Gemma 2 9B
  4. 49.8 BSC-LT/salamandra-7b-instruct
  5. 48.5 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8
  6. 48.3 Qwen3-8B-Base
  7. 48.0 BSC-LT/salamandra-7b
  8. 46.0 meta-llama/Meta-Llama-3.1-8B-Instruct
  9. 44.5 Llama 3.1-8B
  10. 43.7 Gemma 3 4B PT
  11. 42.3 Qwen2.5-7B
  12. 42.2 Gemma 3 4B
  13. 41.9 Qwen2.5-3B
  14. 41.6 Qwen/Qwen2.5-7B-Instruct
  15. 41.5 HiTZ/latxa-7b-v1.2
  16. 41.4 IIC/RigoChat-7b-v2
  17. 39.5 meta-llama/Llama-3.2-3B-Instruct
  18. 38.8 01-ai/Yi-1.5-9B
  19. 38.8 utter-project/EuroLLM-9B-Instruct
  20. 38.0 Qwen/Qwen2.5-3B-Instruct
  21. 37.7 Llama 3.2 3B
  22. 37.6 utter-project/EuroLLM-9B
  23. 37.5 occiglot/occiglot-7b-es-en
  24. 37.5 Qwen2.5-1.5B
  25. 37.1 Aya Expanse 8B
  26. 37.0 Yi-1.5-9B-Chat
  27. 36.8 Gemma 2 2B IT
  28. 36.4 occiglot/occiglot-7b-es-en-instruct
  29. 35.9 mistralai/Mistral-7B-Instruct-v0.3
  30. 35.8 Qwen/Qwen2.5-1.5B-Instruct
  31. 35.7 google/gemma-2-2b
  32. 35.7 occiglot/occiglot-7b-eu5
  33. 35.7 Mistral 7B v0.3
  34. 35.6 occiglot/occiglot-7b-eu5-instruct
  35. 35.6 openGPT-X/Teuken-7B-instruct-commercial-v0.4
  36. 34.8 HuggingFaceTB/SmolLM3-3B
  37. 34.6 HuggingFaceTB/SmolLM3-3B-Base
  38. 34.0 allenai/OLMo-2-1124-7B
  39. 33.7 Qwen3-4B
  40. 33.5 meta-llama/Llama-3.2-1B-Instruct
  41. 33.5 openGPT-X/Teuken-7B-instruct-research-v0.4
  42. 33.4 allenai/OLMo-2-1124-7B-Instruct
  43. 32.9 bertin-project/Gromenauer-7B-Instruct
  44. 32.2 SmolLM2 1.7B Instruct
  45. 31.3 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0
  46. 30.7 Gemma 3 1B
  47. 30.5 Llama 3.2 1B
  48. 30.4 Qwen3-8B
  49. 29.9 SmolLM2-1.7B
  50. 29.9 bertin-project/Gromenauer-7B
  51. 29.9 projecte-aina/aguila-7b
  52. 29.3 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit
  53. 29.1 BSC-LT/salamandra-2b-instruct
  54. 29.1 LenguajeNaturalAI/leniachat-gemma-2b-v0
  55. 29.1 projecte-aina/FLOR-1.3B
  56. 29.1 projecte-aina/FLOR-1.3B-Instructed
  57. 29.1 proxectonos/Carballo-bloom-1.3B
  58. 29.1 utter-project/EuroLLM-1.7B
  59. 29.1 Qwen3-1.7B
  60. 29.0 projecte-aina/FLOR-6.3B
  61. 28.8 projecte-aina/FLOR-6.3B-Instructed
  62. 28.4 Phi-1.5
  63. 27.6 utter-project/EuroLLM-1.7B-Instruct
  64. 27.5 bertin-project/bertin-gpt-j-6B
  65. 27.5 BSC-LT/salamandra-2b
  66. 27.5 DeepSeek-R1-Distill-Qwen-7B
  67. 27.1 Gemma 3 1B PT
  68. 26.5 gplsi/Aitana-6.3B
  69. 26.3 DeepSeek-R1-Distill-Qwen-1.5B
  1. 1 BSC-LT/ALIA-40b-instruct-2601 4-shot 66.2%
  2. 2 google/gemma-2-9b-it 4-shot 51.4%
  3. 3 Gemma 2 9B 4-shot 51%
  4. 4 BSC-LT/salamandra-7b-instruct 4-shot 49.8%
  5. 5 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8 48.5%
  6. 6 Qwen3-8B-Base 48.3%
  7. 7 BSC-LT/salamandra-7b 4-shot 48%
  8. 8 meta-llama/Meta-Llama-3.1-8B-Instruct 4-shot 46%
  9. 9 Llama 3.1-8B 4-shot 44.5%
  10. 10 Gemma 3 4B PT 43.7%
  11. 11 Qwen2.5-7B 4-shot 42.3%
  12. 12 Gemma 3 4B 42.2%
  13. 13 Qwen2.5-3B 41.9%
  14. 14 Qwen/Qwen2.5-7B-Instruct 4-shot 41.6%
  15. 15 HiTZ/latxa-7b-v1.2 4-shot 41.5%
  16. 16 IIC/RigoChat-7b-v2 4-shot 41.4%
  17. 17 meta-llama/Llama-3.2-3B-Instruct 39.5%
  18. 18 01-ai/Yi-1.5-9B 4-shot 38.8%
  19. 19 utter-project/EuroLLM-9B-Instruct 4-shot 38.8%
  20. 20 Qwen/Qwen2.5-3B-Instruct 38%
  21. 21 Llama 3.2 3B 37.7%
  22. 22 utter-project/EuroLLM-9B 4-shot 37.6%
  23. 23 occiglot/occiglot-7b-es-en 4-shot 37.5%
  24. 24 Qwen2.5-1.5B 4-shot 37.5%
  25. 25 Aya Expanse 8B 4-shot 37.1%
  26. 26 Yi-1.5-9B-Chat 4-shot 37%
  27. 27 Gemma 2 2B IT 4-shot 36.8%
  28. 28 occiglot/occiglot-7b-es-en-instruct 4-shot 36.4%
  29. 29 mistralai/Mistral-7B-Instruct-v0.3 4-shot 35.9%
  30. 30 Qwen/Qwen2.5-1.5B-Instruct 4-shot 35.8%
  31. 31 google/gemma-2-2b 4-shot 35.7%
  32. 32 occiglot/occiglot-7b-eu5 4-shot 35.7%
  33. 33 Mistral 7B v0.3 4-shot 35.7%
  34. 34 occiglot/occiglot-7b-eu5-instruct 4-shot 35.6%
  35. 35 openGPT-X/Teuken-7B-instruct-commercial-v0.4 35.6%
  36. 36 HuggingFaceTB/SmolLM3-3B 34.8%
  37. 37 HuggingFaceTB/SmolLM3-3B-Base 34.6%
  38. 38 allenai/OLMo-2-1124-7B 34%
  39. 39 Qwen3-4B 33.7%
  40. 40 meta-llama/Llama-3.2-1B-Instruct 4-shot 33.5%
  41. 41 openGPT-X/Teuken-7B-instruct-research-v0.4 33.5%
  42. 42 allenai/OLMo-2-1124-7B-Instruct 33.4%
  43. 43 bertin-project/Gromenauer-7B-Instruct 4-shot 32.9%
  44. 44 SmolLM2 1.7B Instruct 32.2%
  45. 45 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0 4-shot 31.3%
  46. 46 Gemma 3 1B 30.7%
  47. 47 Llama 3.2 1B 4-shot 30.5%
  48. 48 Qwen3-8B 30.4%
  49. 49 SmolLM2-1.7B 29.9%
  50. 50 bertin-project/Gromenauer-7B 4-shot 29.9%
  51. 51 projecte-aina/aguila-7b 4-shot 29.9%
  52. 52 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit 4-shot 29.3%
  53. 53 BSC-LT/salamandra-2b-instruct 4-shot 29.1%
  54. 54 LenguajeNaturalAI/leniachat-gemma-2b-v0 4-shot 29.1%
  55. 55 projecte-aina/FLOR-1.3B 4-shot 29.1%
  56. 56 projecte-aina/FLOR-1.3B-Instructed 4-shot 29.1%
  57. 57 proxectonos/Carballo-bloom-1.3B 4-shot 29.1%
  58. 58 utter-project/EuroLLM-1.7B 4-shot 29.1%
  59. 59 Qwen3-1.7B 29.1%
  60. 60 projecte-aina/FLOR-6.3B 4-shot 29%
  61. 61 projecte-aina/FLOR-6.3B-Instructed 4-shot 28.8%
  62. 62 Phi-1.5 4-shot 28.4%
  63. 63 utter-project/EuroLLM-1.7B-Instruct 4-shot 27.6%
  64. 64 bertin-project/bertin-gpt-j-6B 4-shot 27.5%
  65. 65 BSC-LT/salamandra-2b 4-shot 27.5%
  66. 66 DeepSeek-R1-Distill-Qwen-7B 4-shot 27.5%
  67. 67 Gemma 3 1B PT 27.1%
  68. 68 gplsi/Aitana-6.3B 4-shot 26.5%
  69. 69 DeepSeek-R1-Distill-Qwen-1.5B 4-shot 26.3%

Source: La Leaderboard’s leaderboard

#EntryScoreSettingDateSource
1 BSC-LT/ALIA-40b-instruct-2601 66.2% 4-shot Official leaderboard
2 google/gemma-2-9b-it 51.4% 4-shot Official leaderboard
3 Gemma 2 9B 51% 4-shot Jun 2024 Official leaderboard
4 BSC-LT/salamandra-7b-instruct 49.8% 4-shot Official leaderboard
5 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8 48.5% Official leaderboard
6 Qwen3-8B-Base 48.3% Apr 2025 Official leaderboard
7 BSC-LT/salamandra-7b 48% 4-shot Official leaderboard
8 meta-llama/Meta-Llama-3.1-8B-Instruct 46% 4-shot Official leaderboard
9 Llama 3.1-8B 44.5% 4-shot Jul 2024 Official leaderboard
10 Gemma 3 4B PT 43.7% Mar 2025 Official leaderboard
11 Qwen2.5-7B 42.3% 4-shot Sep 2024 Official leaderboard
12 Gemma 3 4B 42.2% Mar 2025 Official leaderboard
13 Qwen2.5-3B 41.9% Sep 2024 Official leaderboard
14 Qwen/Qwen2.5-7B-Instruct 41.6% 4-shot Official leaderboard
15 HiTZ/latxa-7b-v1.2 41.5% 4-shot Official leaderboard
16 IIC/RigoChat-7b-v2 41.4% 4-shot Official leaderboard
17 meta-llama/Llama-3.2-3B-Instruct 39.5% Official leaderboard
18 01-ai/Yi-1.5-9B 38.8% 4-shot Official leaderboard
19 utter-project/EuroLLM-9B-Instruct 38.8% 4-shot Official leaderboard
20 Qwen/Qwen2.5-3B-Instruct 38% Official leaderboard
21 Llama 3.2 3B 37.7% Sep 2024 Official leaderboard
22 utter-project/EuroLLM-9B 37.6% 4-shot Official leaderboard
23 occiglot/occiglot-7b-es-en 37.5% 4-shot Official leaderboard
24 Qwen2.5-1.5B 37.5% 4-shot Sep 2024 Official leaderboard
25 Aya Expanse 8B 37.1% 4-shot Oct 2024 Official leaderboard
26 Yi-1.5-9B-Chat 37% 4-shot May 2024 Official leaderboard
27 Gemma 2 2B IT 36.8% 4-shot Jul 2024 Official leaderboard
28 occiglot/occiglot-7b-es-en-instruct 36.4% 4-shot Official leaderboard
29 mistralai/Mistral-7B-Instruct-v0.3 35.9% 4-shot Official leaderboard
30 Qwen/Qwen2.5-1.5B-Instruct 35.8% 4-shot Official leaderboard
31 google/gemma-2-2b 35.7% 4-shot Official leaderboard
32 occiglot/occiglot-7b-eu5 35.7% 4-shot Official leaderboard
33 Mistral 7B v0.3 35.7% 4-shot Sep 2023 Official leaderboard
34 occiglot/occiglot-7b-eu5-instruct 35.6% 4-shot Official leaderboard
35 openGPT-X/Teuken-7B-instruct-commercial-v0.4 35.6% Official leaderboard
36 HuggingFaceTB/SmolLM3-3B 34.8% Official leaderboard
37 HuggingFaceTB/SmolLM3-3B-Base 34.6% Official leaderboard
38 allenai/OLMo-2-1124-7B 34% Official leaderboard
39 Qwen3-4B 33.7% Apr 2025 Official leaderboard
40 meta-llama/Llama-3.2-1B-Instruct 33.5% 4-shot Official leaderboard
41 openGPT-X/Teuken-7B-instruct-research-v0.4 33.5% Official leaderboard
42 allenai/OLMo-2-1124-7B-Instruct 33.4% Official leaderboard
43 bertin-project/Gromenauer-7B-Instruct 32.9% 4-shot Official leaderboard
44 SmolLM2 1.7B Instruct 32.2% Oct 2024 Official leaderboard
45 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0 31.3% 4-shot Official leaderboard
46 Gemma 3 1B 30.7% Mar 2025 Official leaderboard
47 Llama 3.2 1B 30.5% 4-shot Sep 2024 Official leaderboard
48 Qwen3-8B 30.4% Apr 2025 Official leaderboard
49 SmolLM2-1.7B 29.9% Oct 2024 Official leaderboard
50 bertin-project/Gromenauer-7B 29.9% 4-shot Official leaderboard
51 projecte-aina/aguila-7b 29.9% 4-shot Official leaderboard
52 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit 29.3% 4-shot Official leaderboard
53 BSC-LT/salamandra-2b-instruct 29.1% 4-shot Official leaderboard
54 LenguajeNaturalAI/leniachat-gemma-2b-v0 29.1% 4-shot Official leaderboard
55 projecte-aina/FLOR-1.3B 29.1% 4-shot Official leaderboard
56 projecte-aina/FLOR-1.3B-Instructed 29.1% 4-shot Official leaderboard
57 proxectonos/Carballo-bloom-1.3B 29.1% 4-shot Official leaderboard
58 utter-project/EuroLLM-1.7B 29.1% 4-shot Official leaderboard
59 Qwen3-1.7B 29.1% Apr 2025 Official leaderboard
60 projecte-aina/FLOR-6.3B 29% 4-shot Official leaderboard
61 projecte-aina/FLOR-6.3B-Instructed 28.8% 4-shot Official leaderboard
62 Phi-1.5 28.4% 4-shot Sep 2023 Official leaderboard
63 utter-project/EuroLLM-1.7B-Instruct 27.6% 4-shot Official leaderboard
64 bertin-project/bertin-gpt-j-6B 27.5% 4-shot Official leaderboard
65 BSC-LT/salamandra-2b 27.5% 4-shot Official leaderboard
66 DeepSeek-R1-Distill-Qwen-7B 27.5% 4-shot Jan 2025 Official leaderboard
67 Gemma 3 1B PT 27.1% Mar 2025 Official leaderboard
68 gplsi/Aitana-6.3B 26.5% 4-shot Official leaderboard
69 DeepSeek-R1-Distill-Qwen-1.5B 26.3% 4-shot Jan 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Gemma 2 9B 51% 4-shot Jun 2024 Official leaderboard
2 Qwen3-8B-Base 48.3% Apr 2025 Official leaderboard
3 Llama 3.1-8B 44.5% 4-shot Jul 2024 Official leaderboard
4 Gemma 3 4B PT 43.7% Mar 2025 Official leaderboard
5 Qwen2.5-7B 42.3% 4-shot Sep 2024 Official leaderboard
6 Gemma 3 4B 42.2% Mar 2025 Official leaderboard
7 Qwen2.5-3B 41.9% Sep 2024 Official leaderboard
8 Llama 3.2 3B 37.7% Sep 2024 Official leaderboard
9 Qwen2.5-1.5B 37.5% 4-shot Sep 2024 Official leaderboard
10 Aya Expanse 8B 37.1% 4-shot Oct 2024 Official leaderboard
11 Yi-1.5-9B-Chat 37% 4-shot May 2024 Official leaderboard
12 Gemma 2 2B IT 36.8% 4-shot Jul 2024 Official leaderboard
13 Mistral 7B v0.3 35.7% 4-shot Sep 2023 Official leaderboard
14 Qwen3-4B 33.7% Apr 2025 Official leaderboard
15 SmolLM2 1.7B Instruct 32.2% Oct 2024 Official leaderboard
16 Gemma 3 1B 30.7% Mar 2025 Official leaderboard
17 Llama 3.2 1B 30.5% 4-shot Sep 2024 Official leaderboard
18 Qwen3-8B 30.4% Apr 2025 Official leaderboard
19 SmolLM2-1.7B 29.9% Oct 2024 Official leaderboard
20 Qwen3-1.7B 29.1% Apr 2025 Official leaderboard
21 Phi-1.5 28.4% 4-shot Sep 2023 Official leaderboard
22 DeepSeek-R1-Distill-Qwen-7B 27.5% 4-shot Jan 2025 Official leaderboard
23 Gemma 3 1B PT 27.1% Mar 2025 Official leaderboard
24 DeepSeek-R1-Distill-Qwen-1.5B 26.3% 4-shot Jan 2025 Official leaderboard

3 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.

Mistral AI 01.AI Google DeepMind
La Leaderboard leaderboard Not tested since Apr 2025

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.

Plot with other benchmarks
On La Leaderboard’s leaderboard Best so far
0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Gemma 2 9B Yi-1.5-9B-Chat Mistral 7B v0.3 Phi-1.5 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Gemma 2 9B Yi-1.5-9B-Chat

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

All 545 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%, 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.

How we check