beyoru/Rhapsody-4B
Rhapsody-4B is a 4.5 billion parameter language model developed by Beyoru, fine-tuned from Qwen3.5-4B with a 32768-token context length. It is specifically optimized for instruction following across diverse tasks, including roleplay, agentic functions, coding, science, finance, and logical reasoning. This model prioritizes these instruction types to enhance its ability to follow complex directives while maintaining the base model's general capabilities.
Loading preview...
Overview
Rhapsody-4B is a 4.5 billion parameter language model developed by Beyoru, built upon the Qwen3.5-4B architecture. It features a substantial context length of 32768 tokens, enabling it to process and generate longer sequences of text. The model has undergone fine-tuning using a curated mixture of seven distinct instruction datasets.
Key Capabilities
- Enhanced Instruction Following: The model is specifically trained to improve its ability to understand and execute complex instructions.
- Diverse Task Proficiency: Training data prioritizes a range of tasks, including:
- Roleplay scenarios
- Agentic functions
- Code generation and understanding
- Scientific reasoning
- Financial analysis
- Logical problem-solving
- Preserved General Abilities: While specialized for instruction following, Rhapsody-4B retains the broad general capabilities of its base model, Qwen3.5-4B.
When to Use This Model
Rhapsody-4B is particularly well-suited for applications requiring robust instruction following across varied domains. Developers should consider this model for use cases that involve:
- Complex conversational agents or role-playing applications.
- Tasks demanding precise code generation or scientific inquiry.
- Applications in finance or logical reasoning where accurate instruction execution is critical.