fyrenvlad/sucree-sft-v1 is a 14 billion parameter Qwen3-based causal language model developed by fyrenvlad, fine-tuned using Unsloth and Huggingface's TRL library. This model leverages efficient training techniques to achieve faster finetuning. With a 32K context length, it is suitable for applications requiring substantial input processing.
Loading preview...
Model Overview
fyrenvlad/sucree-sft-v1 is a 14 billion parameter language model based on the Qwen3 architecture, developed by fyrenvlad. It was finetuned using the Unsloth library in conjunction with Huggingface's TRL (Transformer Reinforcement Learning) library, enabling significantly faster training times.
Key Characteristics
- Base Model: Finetuned from
unsloth/qwen3-14b-unsloth-bnb-4bit. - Efficient Training: Utilizes Unsloth for 2x faster finetuning, making it an efficient choice for developers.
- Context Length: Supports a substantial context window of 32,768 tokens.
- License: Released under the Apache-2.0 license.
Use Cases
This model is particularly well-suited for developers looking for a Qwen3-based model that benefits from optimized training processes. Its large context window makes it capable of handling complex tasks requiring extensive input or conversation history. The Apache-2.0 license provides flexibility for various applications.