hadadrjt/Qwen3-0.6B

Installation

# If you are using Google Colab, you can skip this step.

# Universal
pip install torch transformers accelerate safetensors --upgrade

# CPU
pip install torch --index-url https://download.pytorch.org/whl/cpu
pip install transformers accelerate safetensors --upgrade

Script for running this model

# You can run in local with GPU or CPU.
# And you can run via Google Colab, for free.
# Open https://colab.research.google.com
# Connect to GPU T4.

import torch

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
from datetime import datetime

# Model
model_name = "hadadrjt/Qwen3-0.6B"

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)

# Get current date
current_date = datetime.now().strftime("%A, %B %d, %Y, %I:%M %p %Z")

# User input
input = "Who are you?"  # Insert your message here.

# Build chat-style prompt
messages = [
    {
        "role": "system",
        "content": (
            "You are Qwen developed by Alibaba Cloud.\n"
            "Your version is Qwen 3 with 0.6 billion parameters.\n"
            "You have been fine-tuned by Hadad Darajat, to suit specific tasks.\n"
            f"Current date: {current_date}"
        )
    },
    {"role": "user", "content": input}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True
)

# Tokenize input
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# Streaming setup
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

# Generate response with streaming
with torch.inference_mode():
    generated_ids = model.generate(
        **model_inputs,
        max_new_tokens=32768,
        temperature=0.6,
        top_p=0.95,
        min_p=0.03,
        top_k=64,
        repetition_penalty=1.0,
        streamer=streamer
    )

Click here to run with Google Colab.

Changelogs

* 2025-05-08 *
- Initial fine-tuned version released.
- Datasets used:
  - https://huggingface.co/datasets/unsloth/OpenMathReasoning-mini
  - https://huggingface.co/datasets/mlabonne/FineTome-100k

* 2025-05-12 *
- 2nd fine-tuned version released.
- Datasets used:
  - https://huggingface.co/datasets/sequelbox/Titanium2.1-DeepSeek-R1