lookarooka/My-Stock-Base-8B

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jun 24, 2026Architecture:Transformer Featherless Exclusive Cold

My-Stock-Base-8B is an 8 billion parameter language model developed by lookarooka, fine-tuned and converted to GGUF format using Unsloth. This model is optimized for efficient deployment and usage in local environments, leveraging Unsloth's accelerated training and conversion process. It is designed for general text generation tasks, providing a balance of performance and accessibility for various applications.

Loading preview...

My-Stock-Base-8B Overview

My-Stock-Base-8B is an 8 billion parameter language model, developed by lookarooka, that has been fine-tuned and converted into the GGUF format. This model leverages the Unsloth framework for its training and conversion process, which is noted for its speed improvements, claiming to be 2x faster.

Key Characteristics

  • Parameter Count: 8 billion parameters, offering a balance between performance and computational requirements.
  • Format: Provided in GGUF format, specifically My-Stock-Base-8B.Q4_K_M.gguf, making it suitable for local inference with tools like llama-cli.
  • Optimization: Fine-tuned and converted using Unsloth, indicating a focus on efficient deployment and potentially faster inference on compatible hardware.
  • GGUF Compatibility: The model's BOS (Beginning Of Sentence) token behavior has been adjusted to ensure proper functionality within the GGUF ecosystem.

Usage Considerations

This model is particularly well-suited for developers and users looking to run an 8B parameter model locally with optimized performance. Its GGUF format and Unsloth-backed conversion suggest ease of integration into existing llama.cpp based workflows. It is primarily intended for text-only applications, as indicated by the llama-cli example usage.