KeefeBuild/Keefe-Discere-v3.3

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 29, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

KeefeBuild/Keefe-Discere-v3.3 is a 7.6 billion parameter Qwen2-based language model developed by KeefeBuild. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling a 2x faster training process. It is designed for general language tasks, leveraging its efficient training methodology to provide a capable foundation.

Loading preview...

Model Overview

KeefeBuild/Keefe-Discere-v3.3 is a 7.6 billion parameter language model based on the Qwen2 architecture, developed by KeefeBuild. This iteration of the Keefe-Discere series was fine-tuned using a combination of Unsloth and Huggingface's TRL library, which significantly accelerated the training process, achieving a 2x speed improvement.

Key Characteristics

  • Architecture: Qwen2-based, a robust foundation for various NLP tasks.
  • Parameter Count: 7.6 billion parameters, offering a balance between performance and computational efficiency.
  • Training Efficiency: Leverages Unsloth for accelerated fine-tuning, making it a practical choice for developers seeking faster iteration cycles.
  • License: Distributed under the Apache-2.0 license, allowing for broad use and modification.

Use Cases

This model is suitable for a wide range of general-purpose language applications where a moderately sized yet efficiently trained model is beneficial. Its fine-tuning approach suggests potential for rapid adaptation to specific downstream tasks.