bunnycore/Qwen-2.5-7B-Deep-Stock-v4

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:Jan 26, 2025Architecture:Transformer0.0K Warm

bunnycore/Qwen-2.5-7B-Deep-Stock-v4 is a 7.6 billion parameter language model created by bunnycore, built upon the Qwen2.5-7B-Instruct base using the Model Stock merge method. This model integrates several specialized Qwen-2.5-7B variants to enhance overall performance. It is designed for general language tasks, demonstrating a balanced performance across various benchmarks including IFEval, BBH, and MATH Lvl 5.

Loading preview...

Model Overview

bunnycore/Qwen-2.5-7B-Deep-Stock-v4 is a 7.6 billion parameter language model developed by bunnycore. It is constructed using the Model Stock merge method, leveraging Qwen/Qwen2.5-7B-Instruct as its foundational base model. This merge incorporates several other specialized models, including bunnycore/FuseQwQen-7B, bunnycore/Qwen-2.5-7B-R1-Stock, and bunnycore/Qwen-2.5-7B-Deep-Stock-v1, aiming to consolidate and improve capabilities across the merged components.

Key Capabilities & Performance

This model demonstrates a balanced performance profile across a range of benchmarks, as evaluated on the Open LLM Leaderboard. Key results include:

  • Avg. Score: 36.10
  • IFEval (0-Shot): 77.53
  • BBH (3-Shot): 35.91
  • MATH Lvl 5 (4-Shot): 48.94
  • MMLU-PRO (5-shot): 37.13

These metrics suggest a general-purpose model suitable for diverse applications, with notable performance in instruction following and mathematical reasoning tasks.

Use Cases

Given its broad capabilities and the nature of its merge, bunnycore/Qwen-2.5-7B-Deep-Stock-v4 is well-suited for:

  • General text generation and understanding tasks.
  • Applications requiring instruction following.
  • Scenarios benefiting from improved mathematical reasoning.
  • As a robust base for further fine-tuning on specific downstream tasks.