README.md

---
license: apache-2.0
base_model: unsloth/llama-3-8b-Instruct-bnb-4bit
tags:
- text-generation
- lora
- unsloth
- conversational
- fine-tuned
- ucortex
---

# 🤖 UCortex Llama-3 8B Chatbot

Welcome to the official repository of **UCortex Llama-3 8B**. This model is a fine-tuned version of Meta's Llama-3-8B-Instruct, trained using **Unsloth** to achieve faster response speeds and optimized memory usage.

## 👤 Model Details
- **Developed by:** Saiyad Uzaib (@Uzaib52)
- **Model Type:** Causal Language Model (Fine-tuned via LoRA)
- **Base Model:** `unsloth/llama-3-8b-Instruct-bnb-4bit`
- **Language:** Hinglish / English / Hindi

---


##  How to Run UCortex Chatbot in Google Colab (Free T4 GPU)

If you want to chat with this model, open a new [Google Colab](https://colab.research.google.com/) notebook, make sure to change the runtime to **T4 GPU** (*Runtime > Change runtime type > T4 GPU*), and run the following code:

```python
# 1. DEPENDENCY INSTALLATION
import sys
import subprocess

print("Installing Unsloth (Fast-Track Mode)... Please wait ~30-40 seconds.")

# install optimized wheels for Colab to prevent import 
subprocess.run([sys.executable, "-m", "pip", "install", "--quiet", 
                "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git",
                "xformers", "trl", "peft", "transformers", "accelerate", "bitsandbytes"])

import site
import importlib
importlib.invalidate_caches()
site.main()

print("Installation complete! Loading UCortex Model...")

# 2. QUICK LOAD THE MODEL
try:
    from unsloth import FastLanguageModel
except ModuleNotFoundError:
    import os
    sys.path.append('/usr/local/lib/python3.12/dist-packages')
    from unsloth import FastLanguageModel

import torch
from transformers import TextIteratorStreamer
from threading import Thread

model_name = "Uzaib52/ucortex-llama3-8b"
max_seq_length = 2048

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = model_name,
    max_seq_length = max_seq_length,
    load_in_4bit = True,
    device_map = "cuda"
)
FastLanguageModel.for_inference(model)
print("UCortex Chatbot is active! Type 'exit' to stop.\n" + "="*50)

# 3. DIRECT CONVERSATION LOOP
prompt_format = "<|begin_of_text|><|start_header_id|>user<|end_header_id|>\n\n{}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"

while True:
    user_input = input("\n You: ")
    if user_input.lower() in ["exit", "quit"]:
        print(" UCortex: Bye!")
        break
    if not user_input.strip():
        continue
        
    formatted_prompt = prompt_format.format(user_input)
    inputs = tokenizer([formatted_prompt], return_tensors="pt").to("cuda")
    streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
    
    generation_kwargs = dict(inputs, streamer=streamer, max_new_tokens=512, use_cache=True)
    thread = Thread(target=model.generate, kwargs=generation_kwargs)
    thread.start()
    
    print("UCortex: ", end="", flush=True)
    for new_text in streamer:
        print(new_text, end="", flush=True)
    print()