tayaee/Qwen2.5-1.5B-Korean-GRPO-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-GRPO-smoke model is an instruction-tuned 1.54 billion parameter causal language model from the Qwen2.5 series, developed by Qwen Team. It features a 32,768 token context length and is optimized for enhanced coding, mathematics, instruction following, and long text generation. This model excels at understanding structured data and generating structured outputs like JSON, with strong multilingual support including Korean.

Loading preview...

Model Overview

This model, tayaee/Qwen2.5-1.5B-Korean-GRPO-smoke, is an instruction-tuned variant of the Qwen2.5 series, developed by the Qwen Team. It is a 1.54 billion parameter causal language model built on a transformer architecture, featuring RoPE, SwiGLU, RMSNorm, and attention QKV bias. It supports a substantial context length of 32,768 tokens and can generate up to 8,192 tokens.

Key Capabilities & Improvements

Qwen2.5 models, including this 1.5B instruction-tuned version, offer significant enhancements over previous iterations:

  • Expanded Knowledge & Skills: Greatly improved capabilities in coding and mathematics due to specialized expert models.
  • Instruction Following: Enhanced ability to follow instructions and generate long texts (over 8K tokens).
  • Structured Data Handling: Better understanding of structured data (e.g., tables) and generation of structured outputs, particularly JSON.
  • Robustness: More resilient to diverse system prompts, improving role-play and chatbot condition-setting.
  • Multilingual Support: Comprehensive support for over 29 languages, including Korean, Chinese, English, Japanese, and more.

Architecture Details

  • Parameters: 1.54 billion (1.31 billion non-embedding).
  • Layers: 28.
  • Attention Heads: 12 for Q and 2 for KV (GQA).

Usage

This model is designed for instruction-following tasks and can be easily integrated using the Hugging Face transformers library, with specific guidance for apply_chat_template to handle system and user prompts effectively.

Finetunes
1 models