KeefeBuild/Keefe-Discere-v3.0-Final

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 27, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

KeefeBuild/Keefe-Discere-v3.0-Final is a 7.6 billion parameter Qwen2-based causal language model developed by KeefeBuild. This model was finetuned using Unsloth and Huggingface's TRL library, resulting in a 2x faster training process. It is designed for general language tasks, building upon its predecessor, Keefe-Discere-v3.0-Ultimate.

Loading preview...

Keefe-Discere-v3.0-Final Overview

KeefeBuild/Keefe-Discere-v3.0-Final is a 7.6 billion parameter language model developed by KeefeBuild. It is a finetuned variant of the Qwen2 architecture, building upon the base model KeefeBuild/Keefe-Discere-v3.0-Ultimate.

Key Characteristics

  • Architecture: Based on the Qwen2 model family.
  • Parameter Count: 7.6 billion parameters, offering a balance between performance and computational efficiency.
  • Training Efficiency: This model was trained significantly faster, specifically 2x faster, by leveraging Unsloth and Huggingface's TRL library. This indicates an optimized finetuning process.

Intended Use Cases

This model is suitable for a variety of general language generation and understanding tasks, benefiting from its efficient training methodology. Its finetuned nature suggests improved performance over its base model for specific applications, though the README does not detail specific benchmarks or target applications beyond general language modeling.

Finetunes
1 models