tepirale/Ornith-Agents-A1-3.7-35B-A3B-dare_ties_v3
tepirale/Ornith-Agents-A1-3.7-35B-A3B-dare_ties_v3 is a 35.1 billion parameter language model merged using the DARE TIES method, based on Qwen/Qwen3.6-35B-A3B. This model integrates capabilities from deepreinforce-ai/Ornith-1.0-35B and InternScience/Agents-A1, leveraging a 32768 token context length. It is designed to combine the strengths of its constituent models, making it suitable for tasks benefiting from a blend of their respective specializations.
Loading preview...
Model Overview
tepirale/Ornith-Agents-A1-3.7-35B-A3B-dare_ties_v3 is a 35.1 billion parameter language model created through a sophisticated merging process. It utilizes the DARE TIES merge method, a technique designed to combine the strengths of multiple pre-trained models efficiently. The base model for this merge is Qwen/Qwen3.6-35B-A3B, providing a robust foundation.
Key Components and Merge Details
This model is a composite of several powerful language models, carefully selected to enhance its overall capabilities. The primary models integrated into this merge include:
- deepreinforce-ai/Ornith-1.0-35B: Contributes its specific pre-training and architectural advantages.
- InternScience/Agents-A1: Adds its unique features, likely related to agentic capabilities or specialized reasoning.
The merging process involved specific configurations for each component, including density and weight parameters, to optimize the blend of their characteristics. The model was processed with a bfloat16 data type, indicating a balance between performance and memory efficiency.
Potential Use Cases
Given its merged nature and the diverse origins of its components, this model is likely well-suited for applications requiring a combination of general language understanding and specialized functionalities inherited from its constituent models. Its 32768 token context length further supports complex tasks requiring extensive contextual awareness.