YOYO-AI/Qwen2.5-14B-YOYO-V4-p2

TEXT GENERATIONConcurrency Cost:1Model Size:14.8BQuant:FP8Ctx Length:32kPublished:Mar 1, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

YOYO-AI/Qwen2.5-14B-YOYO-V4-p2 is a 14.8 billion parameter preview model from the fourth generation of the Qwen-YOYO series, developed by YOYO-AI. This model, with a 32768 token context length, is part of a series exploring distinct merging methodologies to identify the best-performing variant. It is designed as a precursor to a larger official release that will support a 1 million-token context length, focusing on advanced language understanding and generation.

Loading preview...

Qwen2.5-14B-YOYO-V4-p2: Fourth-Generation Preview

This model, Qwen2.5-14B-YOYO-V4-p2, is a 14.8 billion parameter preview version within the fourth generation of the Qwen-YOYO series, developed by YOYO-AI. The "p" in its name signifies its status as a preview, indicating it's an early release in a series of models.

Key Characteristics

  • Parameter Count: 14.8 billion parameters.
  • Context Length: Supports a 32768 token context window.
  • Development Approach: It is one of three planned preview versions, each utilizing distinct merging methodologies during its development.
  • Future Development: The best-performing model from these previews will be selected and further expanded to support an impressive 1 million-token context length for its official release.

Purpose and Use

This model serves as an important step in the development of the next-generation Qwen-YOYO series, allowing for evaluation and refinement of different merging strategies. Developers can explore its capabilities as a robust language model, understanding that it represents a preview of an upcoming, more powerful official release. It is particularly relevant for those interested in the evolution of large language models and advanced context handling.