Parum-Lucis/gemma3-270m-bpe-2K-20y-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-20y-merged is a 0.3 billion parameter language model based on the Google Gemma-3-270m architecture, created by Parum-Lucis using the Task Arithmetic merge method. This model integrates specific capabilities from 'rafurafu/gemma-3-270m-cpt-bpe-2K-20y' into the Gemma base. It is designed for applications requiring a compact yet capable model, leveraging its 32768 token context length for processing longer sequences.

Loading preview...

Model Overview

Parum-Lucis/gemma3-270m-bpe-2K-20y-merged is a compact 0.3 billion parameter language model derived from the google/gemma-3-270m base. It was developed by Parum-Lucis through a model merging process using MergeKit.

Key Characteristics

  • Base Architecture: Utilizes the google/gemma-3-270m as its foundational model.
  • Merge Method: Employs the Task Arithmetic method to combine model weights.
  • Integrated Capabilities: Incorporates specific features from rafurafu/gemma-3-270m-cpt-bpe-2K-20y to enhance its performance.
  • Context Length: Supports a substantial context window of 32768 tokens, enabling the processing of extensive inputs.

Use Cases

This model is suitable for applications where a smaller footprint is critical, but a reasonable context understanding is still required. Its merged nature suggests potential for specialized tasks, though specific performance metrics are not detailed in the provided information. Developers can leverage its 0.3B parameters for efficient deployment in resource-constrained environments or for fine-tuning on domain-specific tasks.