gjyotin305/Qwen2.5-3B-Instruct_old_sft
Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:3.1BQuant:BF16Ctx Length:32kPublished:Jan 3, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

gjyotin305/Qwen2.5-3B-Instruct_old_sft is a 3.1 billion parameter instruction-tuned causal language model developed by gjyotin305. It is finetuned from unsloth/Qwen2.5-3B-Instruct and was trained using Unsloth and Huggingface's TRL library, enabling 2x faster training. This model is optimized for general instruction-following tasks, leveraging its efficient training methodology.

Loading preview...

Model Overview

gjyotin305/Qwen2.5-3B-Instruct_old_sft is a 3.1 billion parameter instruction-tuned language model. It was developed by gjyotin305 and is based on the Qwen2.5-3B-Instruct architecture, specifically finetuned from the unsloth/Qwen2.5-3B-Instruct model.

Key Characteristics

  • Efficient Training: This model was trained with a focus on efficiency, utilizing the Unsloth library in conjunction with Huggingface's TRL library. This combination allowed for a reported 2x faster training process compared to standard methods.
  • Instruction-Tuned: As an instruction-tuned model, it is designed to follow user prompts and instructions effectively, making it suitable for a variety of conversational and task-oriented applications.
  • Apache-2.0 License: The model is released under the Apache-2.0 license, providing broad permissions for use, modification, and distribution.

Use Cases

This model is well-suited for applications requiring a compact yet capable instruction-following language model, particularly where training efficiency is a consideration. Its instruction-tuned nature makes it adaptable for tasks such as:

  • General conversational AI
  • Text generation based on prompts
  • Simple question answering
  • Prototyping and development where resource efficiency is important