hyokwan/gemma2b_test
TEXT GENERATIONConcurrency Cost:1Model Size:2.5BQuant:BF16Ctx Length:8kPublished:Jan 12, 2026License:apache-2.0Architecture:Transformer Open Weights Warm
hyokwan/gemma2b_test is a 2.5 billion parameter causal language model, fine-tuned from Google's Gemma-3-4b-it base model. It is specifically optimized for text generation in Korean, leveraging the hyokwan/test_data dataset. This model is primarily designed for applications requiring accurate and contextually relevant Korean language generation, making it suitable for tasks like content creation or conversational AI in Korean.
Loading preview...
hyokwan/gemma2b_test: A Korean-Optimized Gemma Variant
hyokwan/gemma2b_test is a 2.5 billion parameter language model derived from Google's Gemma-3-4b-it. This model has been specifically fine-tuned for Korean language generation, utilizing the hyokwan/test_data dataset to enhance its performance and contextual understanding in Korean.
Key Capabilities
- Korean Text Generation: Excels at producing coherent and contextually appropriate text in Korean.
- Fine-tuned Performance: Benefits from targeted training on Korean-specific data, improving relevance for local applications.
- Apache 2.0 License: Offers flexibility for both research and commercial use.
Good for
- Korean Content Creation: Generating articles, summaries, or creative text in Korean.
- Korean Conversational AI: Developing chatbots or virtual assistants that interact in Korean.
- Language-Specific Applications: Projects requiring a robust and efficient Korean language model.