Knowledge & Reasoning on ARC-Challenge, ENEM, BLUEX, OAB Exams, BELEBELE, MMLU, GSM8K-PT
56.22K&R Score (NPM)Tucano2-qwen-3.7B-Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Tucano2-qwen-3.7B-InstructVariant=Instruct, Parameters=3.7B2026.03 | 56.22 | |
| Jurema-7BVariant=Instruct, Parameters=7B2026.03 | 50.66 | |
| Qwen2.5-3B-InstructVariant=Instruct, Parameters=3B2026.03 | 47.34 | |
| Gemma-3-Gaia-PT-BR-4b-itVariant=Instruct, Parameters=4B2026.03 | 45 | |
| SmolLM3-3BVariant=Instruct, Parameters=3B2026.03 | 43.99 | |
| Llama-3.2-3B-InstructVariant=Instruct, Parameters=3B2026.03 | 43.08 | |
| Qwen3-4BVariant=Instruct, Parameters=4B2026.03 | 42.33 | |
| Qwen2.5-1.5B-InstructVariant=Instruct, Parameters=1.5B2026.03 | 40.25 | |
| Tucano2-qwen-1.5B-InstructVariant=Instruct, Parameters=1.5B2026.03 | 39.61 | |
| Qwen3-1.7BVariant=Instruct, Parameters=1.7B2026.03 | 28.24 | |
| Tucano2-qwen-0.5B-InstructVariant=Instruct, Parameters=0.5B2026.03 | 27.77 | |
| Llama-3.2-1B-InstructVariant=Instruct, Parameters=1B2026.03 | 15.37 | |
| Qwen3-0.6BVariant=Instruct, Parameters=0.6B2026.03 | 15.13 | |
| Qwen2.5-0.5B-InstructVariant=Instruct, Parameters=0.5B2026.03 | 14.98 |