WithHaebom/hey-make-v3
VISIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:5.1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Oct 6, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
The WithHaebom/hey-make-v3 is a 5.1 billion parameter language model developed by WithHaebom, fine-tuned from unsloth/gemma-4-E2B-it-unsloth-bnb-4bit. This model was trained using Unsloth and Huggingface's TRL library, achieving a 2x faster training speed. It is designed for general language generation tasks, leveraging its efficient training methodology.
Loading preview...
Overview
WithHaebom/hey-make-v3 is a 5.1 billion parameter language model developed by WithHaebom. It is fine-tuned from the unsloth/gemma-4-E2B-it-unsloth-bnb-4bit base model, utilizing an efficient training process. The model was trained 2x faster by incorporating the Unsloth library alongside Huggingface's TRL library, which optimizes the fine-tuning process.
Key Capabilities
- Efficiently Trained: Benefits from a 2x faster training speed due to the integration of Unsloth and Huggingface's TRL library.
- Gemma-4 Base: Built upon the Gemma-4 architecture, providing a strong foundation for language understanding and generation.
- General Purpose: Suitable for a variety of natural language processing tasks.
Good for
- Developers seeking a moderately sized language model (5.1B parameters) with a 32768 token context length.
- Applications requiring a model fine-tuned with efficient training methodologies.
- Experimentation with models derived from the Gemma-4 family.