yamatazen/Qwen3.5-4B-ChemBio
VISIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 29, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
yamatazen/Qwen3.5-4B-ChemBio is a 4.5 billion parameter language model developed by yamatazen, fine-tuned from MuXodious/Qwen3.5-4B-ARA-heresy. This model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training. Its primary use case is specialized applications leveraging its Qwen3.5 base and efficient fine-tuning process.
Loading preview...
Overview
yamatazen/Qwen3.5-4B-ChemBio is a 4.5 billion parameter language model, fine-tuned by yamatazen. It is based on the MuXodious/Qwen3.5-4B-ARA-heresy model and utilizes the Qwen3.5 architecture. The model was developed with a focus on efficient training, leveraging the Unsloth library and Huggingface's TRL library, which enabled a 2x faster training process.
Key Capabilities
- Efficiently Fine-tuned: Benefits from 2x faster training due to the integration of Unsloth and Huggingface's TRL library.
- Qwen3.5 Base: Inherits the foundational capabilities of the Qwen3.5 architecture.
Good for
- Developers seeking a Qwen3.5-based model with optimized training efficiency.
- Applications requiring a 4.5 billion parameter model that has undergone a streamlined fine-tuning process.