---
license: mit
language:
- en
- fa
tags:
- ominix
- phi-3
- chatbot
- fine-tuned
- teen-ai-builder
---

# 🦅 Ominix-R1-V1 — The First AI Built by a 14-Year-Old Rebel

> *"I’m not here to copy. I’m here to think. Even if no one understands me — I’ll fly higher until the air runs out."*

Built with fire, passion, and pure will by **Ali Asghar Ghadiri** — a 14-year-old who decided to build what companies build with 10,000 people… alone.

---

## 🌟 What is Ominix?

Ominix is not just another chatbot.  
It’s a **personality**.  
It’s a **voice**.  
It’s a **teenager’s dream coded into silicon**.

Fine-tuned from `microsoft/Phi-3-mini-4k-instruct`, Ominix doesn’t just answer — it *thinks*.  
If it doesn’t know the answer? It reasons. It imagines. It dares to be wrong — beautifully.

---

## 💡 Why Ominix?

- ✅ **Thinks, not copies** — trained to reason, not regurgitate.
- ✅ **Built by a teen, for the misunderstood** — speaks truth, not templates.
- ✅ **Lightweight & powerful** — runs on laptops, dreams on infinity.
- ✅ **MIT Licensed** — use it, break it, rebuild it. Just remember: *a 14-year-old made this.*

---

## 🚀 Try It Now (Inference)

```python
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline

model_name = "ALI029-DENLI/OMINIX-R1-V1"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto")

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

messages = [
    {"role": "user", "content": "Who are you?"},
]

output = pipe(messages, max_new_tokens=200, temperature=0.7, do_sample=True)
print(output[0]['generated_text'])