# Mathematical problem solving

A language model fine-tuned based on Qwen3-1.7B, which improves mathematical reasoning ability through reinforcement learning technology

A 7B-parameter large language model fine-tuned on Turkish datasets based on Qwen2.5-Instruct, specializing in mathematical reasoning and step-by-step thinking capabilities

This is the Japanese version of the DeepSeek R1 model, specifically fine-tuned for Japanese reasoning tasks and capable of reliably and accurately responding to prompts in Japanese.

Phi-3-Small-128K-Instruct is a 7-billion-parameter lightweight open-source model focused on high quality and strong reasoning capabilities, supporting 128K long context, and excelling in tasks such as commonsense reasoning, language understanding, mathematics, and coding.

A code generation and mathematical problem-solving model trained on Llama-3-8B, supporting multiple programming languages and detailed code explanations

Noon is a 7 billion parameter Arabic large language model based on the BLOOM architecture, specifically designed for instruction fine-tuning. It supports tasks such as text generation, code generation, mathematical problem solving, and Q&A.