Model Overview

WestKunai-XD-7b is a 7 billion parameter language model developed by seyf1elislam, created through a merge of pre-trained models using the mergekit tool and the DARE TIES method. Its foundation is the mistralai/Mistral-7B-v0.1 model, combined with seyf1elislam/WestKunai-Hermes-7b and seyf1elislam/KuTrix-7b.

Key Capabilities & Performance

This model demonstrates solid performance across a range of benchmarks, as evaluated on the Hugging Face Open LLM Leaderboard:

• Average Score: 73.27
• AI2 Reasoning Challenge (25-Shot): 71.25
• HellaSwag (10-Shot): 87.59
• MMLU (5-Shot): 64.69
• TruthfulQA (0-shot): 67.29
• Winogrande (5-shot): 82.24
• GSM8k (5-shot): 66.57

These scores indicate its proficiency in reasoning, common sense, language understanding, and mathematical problem-solving.

Usage & Integration

WestKunai-XD-7b is designed for general text generation tasks. It can be easily integrated into Python applications using the transformers library, with support for bfloat16 dtype and device_map="auto" for efficient deployment. Quantized GGUF versions are also available for optimized inference on various hardware.