Benchmark Almanac

EusReading

BSC-LT/ALIA-40b-instruct-2601 65.6% tops its leaderboard Gemma 2 9B 62.2% leads the models we track
Not in model cards

Models answer comprehension questions about Basque passages from past C1 proficiency examinations. The scorer checks their choices against the exam key.

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

BSC-LT/ALIA-40b-instruct-2601 tops the leaderboard with 65.6% Gemma 2 9B leads with 62.2% 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 62.2%. 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. 65.6 BSC-LT/ALIA-40b-instruct-2601
  2. 62.2 Gemma 2 9B
  3. 59.1 google/gemma-2-9b-it
  4. 50.9 meta-llama/Meta-Llama-3.1-8B-Instruct
  5. 48.0 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8
  6. 47.4 Llama 3.1-8B
  7. 46.9 Qwen3-8B-Base
  8. 44.9 BSC-LT/salamandra-7b-instruct
  9. 38.6 Gemma 3 4B PT
  10. 34.7 Qwen2.5-7B
  11. 34.4 meta-llama/Llama-3.2-3B-Instruct
  12. 33.0 Qwen/Qwen2.5-7B-Instruct
  13. 33.0 Llama 3.2 3B
  14. 32.7 Gemma 3 4B
  15. 32.1 Qwen/Qwen2.5-3B-Instruct
  16. 31.8 IIC/RigoChat-7b-v2
  17. 31.8 Yi-1.5-9B-Chat
  18. 31.8 Aya Expanse 8B
  19. 31.3 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0
  20. 30.7 DeepSeek-R1-Distill-Qwen-7B
  21. 30.4 DeepSeek-R1-Distill-Qwen-1.5B
  22. 29.8 BSC-LT/salamandra-7b
  23. 29.8 Qwen2.5-3B
  24. 29.3 Mistral 7B v0.3
  25. 29.0 occiglot/occiglot-7b-es-en
  26. 29.0 openGPT-X/Teuken-7B-instruct-research-v0.4
  27. 29.0 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit
  28. 28.7 occiglot/occiglot-7b-eu5
  29. 28.4 bertin-project/Gromenauer-7B-Instruct
  30. 28.1 01-ai/Yi-1.5-9B
  31. 28.1 BSC-LT/salamandra-2b-instruct
  32. 28.1 google/gemma-2-2b
  33. 28.1 HuggingFaceTB/SmolLM3-3B-Base
  34. 27.8 openGPT-X/Teuken-7B-instruct-commercial-v0.4
  35. 27.8 Qwen2.5-1.5B
  36. 27.6 bertin-project/Gromenauer-7B
  37. 27.6 utter-project/EuroLLM-9B
  38. 27.6 SmolLM2-1.7B
  39. 27.3 HiTZ/latxa-7b-v1.2
  40. 27.0 allenai/OLMo-2-1124-7B-Instruct
  41. 27.0 HuggingFaceTB/SmolLM3-3B
  42. 27.0 Qwen/Qwen2.5-1.5B-Instruct
  43. 27.0 utter-project/EuroLLM-9B-Instruct
  44. 27.0 SmolLM2 1.7B Instruct
  45. 26.7 meta-llama/Llama-3.2-1B-Instruct
  46. 26.4 projecte-aina/FLOR-1.3B-Instructed
  47. 26.4 utter-project/EuroLLM-1.7B
  48. 26.4 Gemma 2 2B IT
  49. 26.1 occiglot/occiglot-7b-es-en-instruct
  50. 26.1 projecte-aina/aguila-7b
  51. 26.1 projecte-aina/FLOR-1.3B
  52. 26.1 Llama 3.2 1B
  53. 25.9 gplsi/Aitana-6.3B
  54. 25.9 utter-project/EuroLLM-1.7B-Instruct
  55. 25.6 allenai/OLMo-2-1124-7B
  56. 25.6 projecte-aina/FLOR-6.3B
  57. 25.6 Phi-1.5
  58. 25.6 Gemma 3 1B PT
  59. 25.3 projecte-aina/FLOR-6.3B-Instructed
  60. 24.7 BSC-LT/salamandra-2b
  61. 24.7 LenguajeNaturalAI/leniachat-gemma-2b-v0
  62. 24.7 proxectonos/Carballo-bloom-1.3B
  63. 24.4 bertin-project/bertin-gpt-j-6B
  64. 24.4 mistralai/Mistral-7B-Instruct-v0.3
  65. 23.9 Qwen3-8B
  66. 23.6 occiglot/occiglot-7b-eu5-instruct
  67. 23.6 Gemma 3 1B
  68. 23.6 Qwen3-1.7B
  69. 23.6 Qwen3-4B
  1. 1 BSC-LT/ALIA-40b-instruct-2601 1-shot 65.6%
  2. 2 Gemma 2 9B 1-shot 62.2%
  3. 3 google/gemma-2-9b-it 1-shot 59.1%
  4. 4 meta-llama/Meta-Llama-3.1-8B-Instruct 1-shot 50.9%
  5. 5 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8 48%
  6. 6 Llama 3.1-8B 1-shot 47.4%
  7. 7 Qwen3-8B-Base 46.9%
  8. 8 BSC-LT/salamandra-7b-instruct 1-shot 44.9%
  9. 9 Gemma 3 4B PT 38.6%
  10. 10 Qwen2.5-7B 1-shot 34.7%
  11. 11 meta-llama/Llama-3.2-3B-Instruct 34.4%
  12. 12 Qwen/Qwen2.5-7B-Instruct 1-shot 33%
  13. 13 Llama 3.2 3B 33%
  14. 14 Gemma 3 4B 32.7%
  15. 15 Qwen/Qwen2.5-3B-Instruct 32.1%
  16. 16 IIC/RigoChat-7b-v2 1-shot 31.8%
  17. 17 Yi-1.5-9B-Chat 1-shot 31.8%
  18. 18 Aya Expanse 8B 1-shot 31.8%
  19. 19 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0 1-shot 31.3%
  20. 20 DeepSeek-R1-Distill-Qwen-7B 1-shot 30.7%
  21. 21 DeepSeek-R1-Distill-Qwen-1.5B 1-shot 30.4%
  22. 22 BSC-LT/salamandra-7b 1-shot 29.8%
  23. 23 Qwen2.5-3B 29.8%
  24. 24 Mistral 7B v0.3 1-shot 29.3%
  25. 25 occiglot/occiglot-7b-es-en 1-shot 29%
  26. 26 openGPT-X/Teuken-7B-instruct-research-v0.4 29%
  27. 27 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit 1-shot 29%
  28. 28 occiglot/occiglot-7b-eu5 1-shot 28.7%
  29. 29 bertin-project/Gromenauer-7B-Instruct 1-shot 28.4%
  30. 30 01-ai/Yi-1.5-9B 1-shot 28.1%
  31. 31 BSC-LT/salamandra-2b-instruct 1-shot 28.1%
  32. 32 google/gemma-2-2b 1-shot 28.1%
  33. 33 HuggingFaceTB/SmolLM3-3B-Base 28.1%
  34. 34 openGPT-X/Teuken-7B-instruct-commercial-v0.4 27.8%
  35. 35 Qwen2.5-1.5B 1-shot 27.8%
  36. 36 bertin-project/Gromenauer-7B 1-shot 27.6%
  37. 37 utter-project/EuroLLM-9B 1-shot 27.6%
  38. 38 SmolLM2-1.7B 27.6%
  39. 39 HiTZ/latxa-7b-v1.2 1-shot 27.3%
  40. 40 allenai/OLMo-2-1124-7B-Instruct 27%
  41. 41 HuggingFaceTB/SmolLM3-3B 27%
  42. 42 Qwen/Qwen2.5-1.5B-Instruct 1-shot 27%
  43. 43 utter-project/EuroLLM-9B-Instruct 1-shot 27%
  44. 44 SmolLM2 1.7B Instruct 27%
  45. 45 meta-llama/Llama-3.2-1B-Instruct 1-shot 26.7%
  46. 46 projecte-aina/FLOR-1.3B-Instructed 1-shot 26.4%
  47. 47 utter-project/EuroLLM-1.7B 1-shot 26.4%
  48. 48 Gemma 2 2B IT 1-shot 26.4%
  49. 49 occiglot/occiglot-7b-es-en-instruct 1-shot 26.1%
  50. 50 projecte-aina/aguila-7b 1-shot 26.1%
  51. 51 projecte-aina/FLOR-1.3B 1-shot 26.1%
  52. 52 Llama 3.2 1B 1-shot 26.1%
  53. 53 gplsi/Aitana-6.3B 1-shot 25.9%
  54. 54 utter-project/EuroLLM-1.7B-Instruct 1-shot 25.9%
  55. 55 allenai/OLMo-2-1124-7B 25.6%
  56. 56 projecte-aina/FLOR-6.3B 1-shot 25.6%
  57. 57 Phi-1.5 1-shot 25.6%
  58. 58 Gemma 3 1B PT 25.6%
  59. 59 projecte-aina/FLOR-6.3B-Instructed 1-shot 25.3%
  60. 60 BSC-LT/salamandra-2b 1-shot 24.7%
  61. 61 LenguajeNaturalAI/leniachat-gemma-2b-v0 1-shot 24.7%
  62. 62 proxectonos/Carballo-bloom-1.3B 1-shot 24.7%
  63. 63 bertin-project/bertin-gpt-j-6B 1-shot 24.4%
  64. 64 mistralai/Mistral-7B-Instruct-v0.3 1-shot 24.4%
  65. 65 Qwen3-8B 23.9%
  66. 66 occiglot/occiglot-7b-eu5-instruct 1-shot 23.6%
  67. 67 Gemma 3 1B 23.6%
  68. 68 Qwen3-1.7B 23.6%
  69. 69 Qwen3-4B 23.6%

Source: La Leaderboard’s leaderboard

#EntryScoreSettingDateSource
1 BSC-LT/ALIA-40b-instruct-2601 65.6% 1-shot Official leaderboard
2 Gemma 2 9B 62.2% 1-shot Jun 2024 Official leaderboard
3 google/gemma-2-9b-it 59.1% 1-shot Official leaderboard
4 meta-llama/Meta-Llama-3.1-8B-Instruct 50.9% 1-shot Official leaderboard
5 Qwen/Qwen2.5-14B-Instruct-GPTQ-Int8 48% Official leaderboard
6 Llama 3.1-8B 47.4% 1-shot Jul 2024 Official leaderboard
7 Qwen3-8B-Base 46.9% Apr 2025 Official leaderboard
8 BSC-LT/salamandra-7b-instruct 44.9% 1-shot Official leaderboard
9 Gemma 3 4B PT 38.6% Mar 2025 Official leaderboard
10 Qwen2.5-7B 34.7% 1-shot Sep 2024 Official leaderboard
11 meta-llama/Llama-3.2-3B-Instruct 34.4% Official leaderboard
12 Qwen/Qwen2.5-7B-Instruct 33% 1-shot Official leaderboard
13 Llama 3.2 3B 33% Sep 2024 Official leaderboard
14 Gemma 3 4B 32.7% Mar 2025 Official leaderboard
15 Qwen/Qwen2.5-3B-Instruct 32.1% Official leaderboard
16 IIC/RigoChat-7b-v2 31.8% 1-shot Official leaderboard
17 Yi-1.5-9B-Chat 31.8% 1-shot May 2024 Official leaderboard
18 Aya Expanse 8B 31.8% 1-shot Oct 2024 Official leaderboard
19 LenguajeNaturalAI/leniachat-qwen2-1.5B-v0 31.3% 1-shot Official leaderboard
20 DeepSeek-R1-Distill-Qwen-7B 30.7% 1-shot Jan 2025 Official leaderboard
21 DeepSeek-R1-Distill-Qwen-1.5B 30.4% 1-shot Jan 2025 Official leaderboard
22 BSC-LT/salamandra-7b 29.8% 1-shot Official leaderboard
23 Qwen2.5-3B 29.8% Sep 2024 Official leaderboard
24 Mistral 7B v0.3 29.3% 1-shot Sep 2023 Official leaderboard
25 occiglot/occiglot-7b-es-en 29% 1-shot Official leaderboard
26 openGPT-X/Teuken-7B-instruct-research-v0.4 29% Official leaderboard
27 unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit 29% 1-shot Official leaderboard
28 occiglot/occiglot-7b-eu5 28.7% 1-shot Official leaderboard
29 bertin-project/Gromenauer-7B-Instruct 28.4% 1-shot Official leaderboard
30 01-ai/Yi-1.5-9B 28.1% 1-shot Official leaderboard
31 BSC-LT/salamandra-2b-instruct 28.1% 1-shot Official leaderboard
32 google/gemma-2-2b 28.1% 1-shot Official leaderboard
33 HuggingFaceTB/SmolLM3-3B-Base 28.1% Official leaderboard
34 openGPT-X/Teuken-7B-instruct-commercial-v0.4 27.8% Official leaderboard
35 Qwen2.5-1.5B 27.8% 1-shot Sep 2024 Official leaderboard
36 bertin-project/Gromenauer-7B 27.6% 1-shot Official leaderboard
37 utter-project/EuroLLM-9B 27.6% 1-shot Official leaderboard
38 SmolLM2-1.7B 27.6% Oct 2024 Official leaderboard
39 HiTZ/latxa-7b-v1.2 27.3% 1-shot Official leaderboard
40 allenai/OLMo-2-1124-7B-Instruct 27% Official leaderboard
41 HuggingFaceTB/SmolLM3-3B 27% Official leaderboard
42 Qwen/Qwen2.5-1.5B-Instruct 27% 1-shot Official leaderboard
43 utter-project/EuroLLM-9B-Instruct 27% 1-shot Official leaderboard
44 SmolLM2 1.7B Instruct 27% Oct 2024 Official leaderboard
45 meta-llama/Llama-3.2-1B-Instruct 26.7% 1-shot Official leaderboard
46 projecte-aina/FLOR-1.3B-Instructed 26.4% 1-shot Official leaderboard
47 utter-project/EuroLLM-1.7B 26.4% 1-shot Official leaderboard
48 Gemma 2 2B IT 26.4% 1-shot Jul 2024 Official leaderboard
49 occiglot/occiglot-7b-es-en-instruct 26.1% 1-shot Official leaderboard
50 projecte-aina/aguila-7b 26.1% 1-shot Official leaderboard
51 projecte-aina/FLOR-1.3B 26.1% 1-shot Official leaderboard
52 Llama 3.2 1B 26.1% 1-shot Sep 2024 Official leaderboard
53 gplsi/Aitana-6.3B 25.9% 1-shot Official leaderboard
54 utter-project/EuroLLM-1.7B-Instruct 25.9% 1-shot Official leaderboard
55 allenai/OLMo-2-1124-7B 25.6% Official leaderboard
56 projecte-aina/FLOR-6.3B 25.6% 1-shot Official leaderboard
57 Phi-1.5 25.6% 1-shot Sep 2023 Official leaderboard
58 Gemma 3 1B PT 25.6% Mar 2025 Official leaderboard
59 projecte-aina/FLOR-6.3B-Instructed 25.3% 1-shot Official leaderboard
60 BSC-LT/salamandra-2b 24.7% 1-shot Official leaderboard
61 LenguajeNaturalAI/leniachat-gemma-2b-v0 24.7% 1-shot Official leaderboard
62 proxectonos/Carballo-bloom-1.3B 24.7% 1-shot Official leaderboard
63 bertin-project/bertin-gpt-j-6B 24.4% 1-shot Official leaderboard
64 mistralai/Mistral-7B-Instruct-v0.3 24.4% 1-shot Official leaderboard
65 Qwen3-8B 23.9% Apr 2025 Official leaderboard
66 occiglot/occiglot-7b-eu5-instruct 23.6% 1-shot Official leaderboard
67 Gemma 3 1B 23.6% Mar 2025 Official leaderboard
68 Qwen3-1.7B 23.6% Apr 2025 Official leaderboard
69 Qwen3-4B 23.6% Apr 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Gemma 2 9B 62.2% 1-shot Jun 2024 Official leaderboard
2 Llama 3.1-8B 47.4% 1-shot Jul 2024 Official leaderboard
3 Qwen3-8B-Base 46.9% Apr 2025 Official leaderboard
4 Gemma 3 4B PT 38.6% Mar 2025 Official leaderboard
5 Qwen2.5-7B 34.7% 1-shot Sep 2024 Official leaderboard
6 Llama 3.2 3B 33% Sep 2024 Official leaderboard
7 Gemma 3 4B 32.7% Mar 2025 Official leaderboard
8 Yi-1.5-9B-Chat 31.8% 1-shot May 2024 Official leaderboard
9 Aya Expanse 8B 31.8% 1-shot Oct 2024 Official leaderboard
10 DeepSeek-R1-Distill-Qwen-7B 30.7% 1-shot Jan 2025 Official leaderboard
11 DeepSeek-R1-Distill-Qwen-1.5B 30.4% 1-shot Jan 2025 Official leaderboard
12 Qwen2.5-3B 29.8% Sep 2024 Official leaderboard
13 Mistral 7B v0.3 29.3% 1-shot Sep 2023 Official leaderboard
14 Qwen2.5-1.5B 27.8% 1-shot Sep 2024 Official leaderboard
15 SmolLM2-1.7B 27.6% Oct 2024 Official leaderboard
16 SmolLM2 1.7B Instruct 27% Oct 2024 Official leaderboard
17 Gemma 2 2B IT 26.4% 1-shot Jul 2024 Official leaderboard
18 Llama 3.2 1B 26.1% 1-shot Sep 2024 Official leaderboard
19 Phi-1.5 25.6% 1-shot Sep 2023 Official leaderboard
20 Gemma 3 1B PT 25.6% Mar 2025 Official leaderboard
21 Qwen3-8B 23.9% Apr 2025 Official leaderboard
22 Gemma 3 1B 23.6% Mar 2025 Official leaderboard
23 Qwen3-1.7B 23.6% Apr 2025 Official leaderboard
24 Qwen3-4B 23.6% Apr 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
Gemma 2 9B 62.2%

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 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 62.2%, 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