claude-code-kim/qwen-0.5b-brain-v4

TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 19, 2026Architecture:Transformer Featherless Exclusive Cold

The claude-code-kim/qwen-0.5b-brain-v4 is a 0.5 billion parameter Qwen2.5-instruct model, fine-tuned and converted to GGUF format by claude-code-kim using Unsloth. This compact model is optimized for efficient deployment and inference, supporting a 32768 token context length. It is designed for general instruction-following tasks, leveraging its small size for resource-constrained environments.

Loading preview...

Model Overview

The claude-code-kim/qwen-0.5b-brain-v4 is a compact 0.5 billion parameter Qwen2.5-instruct model, specifically fine-tuned and converted into the GGUF format. This model was processed using Unsloth, a framework known for accelerating training and conversion processes, enabling this model to be trained significantly faster.

Key Characteristics

  • Architecture: Based on the Qwen2.5-instruct series, providing general instruction-following capabilities.
  • Parameter Count: A highly efficient 0.5 billion parameters, making it suitable for edge devices or applications with limited computational resources.
  • Context Length: Supports a substantial context window of 32768 tokens, allowing for processing longer inputs and maintaining conversational coherence.
  • Format: Provided in GGUF format, ensuring broad compatibility with various inference engines like llama-cli and ollama.
  • Optimization: Benefits from Unsloth's optimizations, leading to faster training and efficient deployment.

Deployment and Usage

An Ollama Modelfile is included for straightforward deployment with the Ollama ecosystem. Example command-line usage is provided for both text-only and multimodal llama-cli applications, indicating its readiness for immediate integration into projects requiring a lightweight yet capable instruction-tuned model.