Gunulhona/Gemma-3-27B-Text-Only

TEXT GENERATIONPricing:Input $1.06 / Cached $0.053 / Output $2.6Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kPublished:Dec 9, 2025Architecture:Transformer Featherless Exclusive Cold

Gunulhona/Gemma-3-27B-Text-Only is a 27 billion parameter language model, merged using the Arcee Fusion method. It is based on google/medgemma-27b-text-it, incorporating layers from an additional model. This model is designed for text-only applications, leveraging its substantial parameter count for robust language understanding and generation tasks.

Loading preview...

Model Overview

Gunulhona/Gemma-3-27B-Text-Only is a 27 billion parameter language model created through a merge process using MergeKit. This model leverages the Arcee Fusion merge method, building upon the google/medgemma-27b-text-it as its base.

Merge Details

The model integrates layers from google/medgemma-27b-text-it and an additional component identified as /content/gemma-3-proxy. Specifically, layers 0 through 62 from both source models were combined to form this merged architecture. This approach aims to combine the strengths of its constituent models into a single, more capable entity.

Key Characteristics

  • Parameter Count: 27 billion parameters, indicating a large-scale model suitable for complex language tasks.
  • Base Model: Derived from google/medgemma-27b-text-it, suggesting a foundation in medical or general text-based applications.
  • Merge Method: Utilizes Arcee Fusion for combining model weights, a technique designed to create powerful composite models.
  • Context Length: Supports a context length of 32768 tokens, enabling the processing of extensive inputs and generating coherent long-form text.

Potential Use Cases

Given its text-only nature and substantial parameter count, this model is well-suited for:

  • Advanced text generation and completion.
  • Complex natural language understanding tasks.
  • Applications requiring a broad understanding of language nuances.
  • Research and development in large language model capabilities.