khazarai/StockDirection-6K

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:0.8BQuant:BF16Ctx Length:32kPublished:Mar 11, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

khazarai/StockDirection-6K is a 0.8 billion parameter language model fine-tuned for binary stock movement prediction. This model specializes in forecasting whether the next day's stock price for Akbank (AKBNK) on Borsa Istanbul will move UP or DOWN. It processes formatted prompts detailing the last five days of daily percentage price changes to output a categorical prediction. StockDirection-6K is designed for educational and research purposes in financial classification tasks, not for live trading.

Loading preview...

StockDirection-6K: Specialized Stock Movement Prediction

StockDirection-6K is a fine-tuned language model developed by khazarai, specifically designed for binary stock movement prediction. Unlike general-purpose LLMs, this 0.8 billion parameter model focuses on a very narrow, yet complex, financial task.

Key Capabilities & Features

  • Targeted Prediction: Predicts "UP" or "DOWN" for the next day's stock price of Akbank (AKBNK) on Borsa Istanbul.
  • Input Format: Utilizes a structured prompt detailing the daily percentage price changes over the last five days.
  • Fine-tuned Data: Trained on a dataset of 6,300 labeled rows of historical AKBNK stock data, emphasizing sequence learning for financial time series.

Use Cases & Limitations

This model is primarily intended for:

  • Educational Purposes: Demonstrating the application of LLMs in financial classification.
  • Research: Exploring text-based sequence learning for stock direction.
  • Proof of Concept: Serving as an example for specific stock price direction prediction using natural language prompts.

Important Note: This model is not intended for financial advice or live trading decisions. Its purpose is strictly for academic and experimental exploration of AI in finance.