Parum-Lucis/gemma3-270m-bpe-2K-5y-merged
TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.3BQuant:BF16Context Size:32kPublished:Jul 8, 2026Architecture:Transformer Featherless Exclusive Cold
Parum-Lucis/gemma3-270m-bpe-2K-5y-merged is a 0.3 billion parameter language model based on the Google Gemma-3 architecture, created by Parum-Lucis through a merge of pre-trained models. This model was developed using the Task Arithmetic merge method, combining google/gemma-3-270m with rafurafu/gemma-3-270m-cpt-bpe-2K-5y. It is designed for general language tasks, leveraging the strengths of its merged components.
Loading preview...
Overview
This model, gemma3-270m-bpe-2K-5y-merged, is a 0.3 billion parameter language model derived from the Google Gemma-3 architecture. It was created by Parum-Lucis using the Task Arithmetic merge method, combining google/gemma-3-270m as the base with rafurafu/gemma-3-270m-cpt-bpe-2K-5y.
Key Capabilities
- Merged Architecture: Leverages the Gemma-3 270M parameter base model, enhanced by a specialized component.
- Task Arithmetic Method: Utilizes a specific merging technique to combine the capabilities of its constituent models.
- Compact Size: At 0.3 billion parameters, it offers a balance between performance and computational efficiency.
Good for
- Resource-constrained environments: Its smaller size makes it suitable for deployment where computational resources are limited.
- General language understanding: Can be applied to a variety of common natural language processing tasks.
- Experimentation with merged models: Provides a practical example of a model created via the Task Arithmetic method, useful for researchers and developers exploring model merging techniques.