Benchmark
Estonian Native Benchmark declension accuracy (instruction-tuned models)
Multilingual benchmarks compare model performance across languages. Higher scores are better.
- Models and systems measured
- 26
- Scores recorded
- 26
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 0.582
- Mean
- 0.631
- Median
- 0.279
- Std dev
- 0.274
- 10th percentile
- 0.937
- 90th percentile
- 0.056
- Min
Results
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| Claude Sonnet 3.7instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.937 |
| Claude Sonnet 4instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.822 |
| Deepseek V3 (671B-A37B)instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.631 |
| EuroLLM-9B-Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.644 |
| Gemini 2.0 Flashinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.932 |
| Gemini 2.5 Proinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.955 |
| Gemma-2-27B-Itinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.406 |
| Gemma-2-9B-Itinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.308 |
| Gemma-3 27B Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.404 |
| Gemma-3-12b-Itinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.356 |
| GPT-4oinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.948 |
| GPT-4o Miniinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.554 |
| GPT-5instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.934 |
| GPT-OSS-120Binstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.356 |
| Kimi-K2-0905 (1000B-A32B)instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.876 |
| Llama-3.1 405B Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.68 |
| Llama-3.1 70B Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.464 |
| Llama-3.1 8B EstLLM-0825instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.811 |
| Llama-3.1 8B Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.099 |
| Llama-4 Maverick (400B-A17B)instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.828 |
| Llama-4 Scout (109B-A17B)instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.766 |
| Mistral Nemo Instruct (12B)instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.274 |
| Qwen2-72B-Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.177 |
| Qwen3-235B-A22B-2507instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.555 |
| Qwen3-4B-Instruct-2507instruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.056 |
| Qwen3-Next-80B-A3B-Instructinstruction-tuned modelstable: 9, setting: instruction-tuned models, task: Declension· Estonian Native Benchmark (arXiv:2510.21193v2) data | 0.362 |