Elsephire/Qwen3.6-35B-A3B-Ornith-80-20-Beta

TEXT GENERATIONConcurrent Unit Cost:3Model Size:35.1BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 2, 2026License:mitArchitecture:Transformer0.0K Open Weights Featherless Exclusive Cold

Elsephire/Qwen3.6-35B-A3B-Ornith-80-20-Beta is a 35.1 billion parameter language model created by merging Qwen3.6-35B-A3B and Ornith-1.0-35B using Task Arithmetic with an 80/20 weighting. This beta release, developed with Minerve-Fusion, combines the strengths of its base models, offering a robust foundation for various language-based tasks. It features a 32768 token context length, making it suitable for processing extensive inputs.

Loading preview...

Model Overview

Elsephire/Qwen3.6-35B-A3B-Ornith-80-20-Beta is a 35.1 billion parameter language model, representing a significant advancement in model fusion. It was meticulously constructed by merging two powerful base models: Qwen3.6-35B-A3B and Ornith-1.0-35B. This merge was executed using the Task Arithmetic method, applying an 80/20 weighting ratio to balance the contributions of each constituent model.

Key Characteristics

  • Fusion Technology: Developed using Minerve-Fusion, a tool slated for open-source release, demonstrating innovative model combination techniques.
  • Parameter Count: Boasts 35.1 billion parameters, placing it in the large-scale language model category.
  • Context Length: Supports a substantial context window of 32768 tokens, enabling it to handle complex and lengthy inputs effectively.

Current Status

This model is currently a beta release. Users are encouraged to engage in testing and provide feedback to contribute to its ongoing development and refinement. Its unique merged architecture suggests potential for diverse applications, leveraging the combined capabilities of its foundational models.

Finetunes
2 models