mrzenin/Decka-4B

VISIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 4, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Decka-4B is a 4.5 billion parameter Qwen3.5-based language model developed by mrzenin. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling faster training. It is designed for general language tasks, leveraging its efficient fine-tuning process.

Loading preview...

mrzenin/Decka-4B: An Efficiently Fine-Tuned Qwen3.5 Model

mrzenin/Decka-4B is a 4.5 billion parameter language model built upon the Qwen3.5 architecture. Developed by mrzenin, this model distinguishes itself through its fine-tuning process, which leveraged Unsloth and Huggingface's TRL library. This approach allowed for significantly faster training, making it an efficient option for various language generation and understanding tasks.

Key Capabilities

  • Efficient Training: Fine-tuned with Unsloth, enabling quicker iteration and deployment.
  • Qwen3.5 Base: Benefits from the robust capabilities of the Qwen3.5 foundational model.
  • General Purpose: Suitable for a broad range of natural language processing applications.

Good For

  • Developers seeking a moderately sized model with a strong base.
  • Applications requiring efficient fine-tuning and deployment.
  • General text generation, summarization, and conversational AI tasks.