---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
---

# Gemma2-2B-mixed

Gemma2-2B-mixed is a compact LLM designed for deployment on smartphones. Gemma2-2B-mixed is a novel AI model that combines the strengths of multiple uncensored models to create a powerful and versatile language processing model. By merging a range of models, including a roleplay uncensored model, Gemma2-2B-mixed offers unparalleled capabilities for generating creative and informative responses.

## Key Features
- Uncensored: Gemma2-2B-mixed is designed to provide unfiltered and unbiased responses, making it an ideal choice for applications where creativity and honesty are essential.
- Compact Size: Despite its advanced capabilities, Gemma2-2B-mixed is optimized for deployment on smartphones and other resource-constrained devices.
- Roleplay Capabilities: The incorporation of a roleplay uncensored model enables Gemma2-2B-mixed to engage in immersive and interactive conversations.
- Offline Capabilities: Gemma2-2B-mixed can operate effectively in environments with limited or no internet connectivity.

## Benefits of Uncensored Model
- Increased Creativity: By removing censorship constraints, Gemma2-2B-mixed can generate more innovative and diverse responses.
- Improved Question Answering: The model's uncensored nature allows it to provide more direct and accurate answers to a wide range of questions.
- Enhanced Overall Performance: Gemma2-2B-mixed offers a more comprehensive and engaging user experience, making it an excellent choice for various applications.

Gemma2-2B-mixed is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):

## 🧩 Configuration

```yaml
models:
  - model: IlyaGusev/gemma-2-2b-it-abliterated
  - model: TheDrummer/Gemmasutra-Mini-2B-v1
  - model: monsterapi/gemma-2-2b-LoRA-MonsterInstruct
merge_method: model_stock
base_model: IlyaGusev/gemma-2-2b-it-abliterated
dtype: bfloat16

```