jerrycheng233/model6_gspo_qwen3_16bit
The jerrycheng233/model6_gspo_qwen3_16bit is a 14 billion parameter Qwen3-based causal language model developed by jerrycheng233. It was fine-tuned from TeichAI/Qwen3-14B-Claude-4.5-Opus-High-Reasoning-Distill using Unsloth and Huggingface's TRL library, resulting in 2x faster training. This model is optimized for high-reasoning tasks, leveraging its foundation for advanced cognitive capabilities.
Loading preview...
Model Overview
The jerrycheng233/model6_gspo_qwen3_16bit is a 14 billion parameter language model developed by jerrycheng233. It is based on the Qwen3 architecture and was fine-tuned from the TeichAI/Qwen3-14B-Claude-4.5-Opus-High-Reasoning-Distill model. This fine-tuning process specifically leveraged Unsloth and Huggingface's TRL library, which enabled a 2x faster training speed.
Key Capabilities
- High Reasoning: Inherits and refines the high-reasoning capabilities from its base model, making it suitable for complex analytical tasks.
- Efficient Training: Benefits from the Unsloth framework, indicating an optimized and resource-efficient development process.
- Qwen3 Architecture: Built upon the robust Qwen3 foundation, suggesting strong general language understanding and generation abilities.
Good For
- Applications requiring advanced reasoning and problem-solving.
- Scenarios where efficient model deployment and fine-tuning are beneficial.
- Tasks that can leverage a powerful 14B parameter model with a 32768 token context length.