Shaheen-Qwen2.5-3B-Kulliyat-e-Iqbal: Specialized Iqbaliyat LLM

Khurram123's Shaheen-Qwen2.5-3B-Kulliyat-e-Iqbal is a 3.1 billion parameter Large Language Model, fine-tuned on the Qwen 2.5 3B Instruct architecture. Its core purpose is to provide in-depth understanding and explanation of the philosophical and poetic works of Allama Muhammad Iqbal, covering both his Urdu and Persian writings.

Key Capabilities

• Specialized Knowledge: Deeply trained on 11 major Urdu and Persian books by Allama Iqbal, using the Kulliyat-e-Iqbal Shaheen Dataset (11,659 instruction-response pairs).
• Bilingual Proficiency: Capable of interpreting and explaining Persian (Farsi) couplets in simple Urdu.
• Philosophical Insight: Optimized to discuss and clarify core Iqbalian concepts such as Khudi (Selfhood), Ishq (Divine Love), and Shaheen (The Eagle).
• Efficiency: A lightweight 3 billion parameter model, fine-tuned with 4-bit LoRA using Unsloth, ensuring fast inference on consumer-grade hardware like the RTX 4060 Ti with minimal VRAM usage.

Good For

• Researchers and students studying Allama Iqbal's philosophy and poetry.
• Applications requiring contextual understanding and explanation of Iqbaliyat in Urdu and Persian.
• Developing conversational AI focused on classical South Asian literature and philosophy.