hiep-2/qwen3-0.6b-math-cpt-sft
TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 14, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
The hiep-2/qwen3-0.6b-math-cpt-sft is a 0.8 billion parameter Qwen3-based language model developed by hiep-2. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is optimized for specific tasks, leveraging its efficient training methodology for focused applications.
Loading preview...
Model Overview
The hiep-2/qwen3-0.6b-math-cpt-sft is a 0.8 billion parameter language model based on the Qwen3 architecture. Developed by hiep-2, this model was fine-tuned from unsloth/qwen3-0.6b-unsloth-bnb-4bit using the Unsloth library and Huggingface's TRL. A key characteristic of its development is the use of Unsloth, which facilitated a 2x faster training process.
Key Capabilities
- Efficient Training: Leverages Unsloth for significantly accelerated fine-tuning.
- Qwen3 Architecture: Built upon the robust Qwen3 foundation.
- Specific Fine-tuning: Designed for focused applications through its fine-tuned nature.
Good For
- Developers seeking a compact Qwen3-based model with efficient training origins.
- Use cases where faster fine-tuning is a critical advantage.
- Applications requiring a specialized model derived from the Qwen3 family.