✨ Jaadhu-1.5B

---

language: en license: other tags: - llm - transformers - generative - jaadhu - qwen datasets: - custom base_model: Qwen/Qwen2.5-1.5B pipeline_tag: text-generation

Jaadhu-1.5B is a general-purpose large language model based on Qwen2.5-1.5B.
The name “Jaadhu” (जादू / جادو) means magic, symbolizing the transformative power of AI in bringing words and ideas to life.

---

🔧 Model Details

• Base model: Qwen2.5-1.5B
• Parameters: ~1.5 billion
• Architecture: Transformer decoder-only
• Fine-tuning method: LoRA adapters (q_proj, v_proj), merged into base
• Framework: Hugging Face Transformers + PEFT

---

🚀 Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "jaadhu-ai/Jaadhu-1.5B"

tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, dtype=torch.float32).to("mps")

prompt = "Write a short story about the magic of monsoon in Kerala."
x = tok(prompt, return_tensors="pt").to(model.device)
y = model.generate(**x, max_new_tokens=100, temperature=0.7, top_p=0.9)
print(tok.decode(y[0], skip_special_tokens=True))

📊 Intended Use

General-purpose natural language understanding and generation

Chatbots, summarization, question answering, creative writing

Lightweight reasoning tasks

⚠️ Limitations

At 1.5B parameters, less capable than larger models

May hallucinate facts or produce biased text

Not safety-aligned for sensitive applications

📜 License

This model inherits the license of Qwen2.5-1.5B
.
Please review licensing terms before commercial use.

✨ Acknowledgements

Base: Qwen2.5-1.5B

Training: Custom Apple Silicon fine-tuning pipeline

Inspiration: “Jaadhu” — meaning magic, representing the wonder of AI