tayaee/Qwen2.5-1.5B-Korean-ORPO-smoke

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 27, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The tayaee/Qwen2.5-1.5B-Korean-ORPO-smoke model is an instruction-tuned causal language model from the Qwen2.5 series, developed by Qwen. With 1.54 billion parameters and a 32,768-token context length, it features significant improvements in coding, mathematics, instruction following, and long text generation. This model is particularly optimized for understanding structured data like JSON and offers robust multilingual support for over 29 languages, including Korean.

Loading preview...

Model Overview

The tayaee/Qwen2.5-1.5B-Korean-ORPO-smoke model is an instruction-tuned variant of the Qwen2.5 series, developed by Qwen. This 1.54 billion parameter causal language model builds upon the Qwen2 architecture, incorporating transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias, and tied word embeddings. It supports a full context length of 32,768 tokens and can generate up to 8,192 tokens.

Key Capabilities

  • Enhanced Knowledge & Reasoning: Significantly improved capabilities in coding and mathematics, leveraging specialized expert models.
  • Instruction Following: Demonstrates substantial improvements in adhering to instructions and generating structured outputs, particularly JSON.
  • Long Text Generation: Excels at generating extended texts, handling outputs over 8,000 tokens.
  • Structured Data Understanding: Better at interpreting structured data formats, such as tables.
  • Multilingual Support: Offers robust support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, and Korean.
  • System Prompt Resilience: More resilient to diverse system prompts, enhancing role-play and chatbot condition-setting.

When to Use This Model

This model is suitable for applications requiring strong instruction following, code generation, mathematical problem-solving, and the ability to process and generate long, structured texts. Its multilingual capabilities make it a strong candidate for global applications, especially those involving Korean language processing. Developers can leverage its improved JSON generation for reliable structured output in various tasks.

Finetunes
1 models