Lunar_10.7B: A Companion-Oriented Language Model

Lunar_10.7B is a 10.7 billion parameter model developed by ChaoticNeutrals, created through a SLERP merge with the Sao10K/Sensualize-Solar-10.7B model. Its primary design goal is to serve as a companion bot, capable of engaging in both intimate and general conversations.

Key Capabilities & Performance

This model has been evaluated on the Open LLM Leaderboard, demonstrating solid performance across a range of benchmarks:

• Average Score: 67.25
• AI2 Reasoning Challenge (25-Shot): 65.87
• HellaSwag (10-Shot): 84.85
• MMLU (5-Shot): 64.23
• TruthfulQA (0-shot): 53.51
• Winogrande (5-shot): 81.37
• GSM8k (5-shot): 53.68

These scores indicate its proficiency in reasoning, common sense, and general knowledge tasks, supporting its role in interactive dialogue systems. The model's GGUF quantization is also available for optimized deployment.

Ideal Use Cases

Lunar_10.7B is particularly well-suited for applications requiring:

• Interactive companion bots: Designed for engaging in personal and intimate conversations.
• Dialogue systems: Where nuanced and context-aware responses are crucial.
• Creative writing assistance: Generating conversational text with emotional depth.
• Role-playing scenarios: Adapting to various conversational styles and personas.