dipta007/GanitLLM-4B-SFT is a 4 billion parameter causal language model, based on Qwen3-4B, specifically fine-tuned for mathematical reasoning in Bengali. It significantly improves accuracy on Bengali math benchmarks (Bn-MGSM and Bn-MSVAMP) and generates more concise solutions compared to its base model. This model is optimized for solving Bengali math problems with step-by-step reasoning.
No reviews yet. Be the first to review!
Hugging Face