Natural Language Understanding and Reasoning on General Benchmarks Italian benchmark leaderboard · SOTA2 Research

Natural Language Understanding and Reasoning on General Benchmarks Italian

37.47ARC-C-itQwen2.5

Evaluation Results

Method	Links
Qwen2.52025.12		37.47	70.44	47.55	46.71	43.33	49.1
Gamayun2025.12		35.9	67.22	52.63	41.87	41.28	47.78
Qwen32025.12		35.33	63.89	45.61	49.13	40.24	46.84
Gemma32025.12		34.99	42.56	47.43	37.19	36.27	39.69
EuroLM2025.12		32.08	24.56	50.75	27.4	33.96	33.75
Llama3.22025.12		29.34	48	42.34	38.64	35.65	38.79