zlyngkhoi/qwen2.5-0.5b-instruction-following-sft

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 18, 2026Architecture:Transformer Featherless Exclusive Cold

The zlyngkhoi/qwen2.5-0.5b-instruction-following-sft is a 0.5 billion parameter causal language model, fine-tuned by zlyngkhoi for instruction following. Built using Aligntune and based on the Qwen2.5-0.5B-Instruct architecture, this model is optimized for processing and responding to instructions. It leverages SFT (Supervised Fine-Tuning) with Unsloth and TRL backends, making it suitable for applications requiring efficient instruction-based text generation.

Loading preview...

Model Overview

This model, zlyngkhoi/qwen2.5-0.5b-instruction-following-sft, is a 0.5 billion parameter instruction-following language model. It was developed by zlyngkhoi and fine-tuned from the Qwen/Qwen2.5-0.5B-Instruct base model. The fine-tuning process utilized Aligntune, a framework supporting various open-source models and algorithms, specifically employing Supervised Fine-Tuning (SFT) for instruction following.

Key Characteristics

  • Base Model: Qwen2.5-0.5B-Instruct
  • Parameter Count: 0.5 billion
  • Context Length: 32768 tokens
  • Fine-tuning Method: SFT (Instruction Following)
  • Backend Technologies: Unsloth + TRL, indicating an optimization for efficient training and inference.

Intended Use Cases

This model is designed for tasks that require adherence to specific instructions, making it suitable for:

  • Generating text based on explicit prompts.
  • Developing lightweight conversational agents.
  • Applications where efficient instruction processing on a smaller model is beneficial.