| Property | Value |
|---|---|
| Parameter Count | 1.54B |
| Model Type | Text Generation (Role-Play) |
| Base Model | Qwen2-1.5B |
| License | Apache-2.0 |
| Max Input Tokens | 32,768 |
| Max Output Tokens | 8,192 |
What is oxy-1-micro?
Oxy-1-micro is a specialized language model developed by Oxygen (oxyapi) with contributions from TornadoSoftwares. It's a fine-tuned version of Qwen2-1.5B, specifically optimized for role-play scenarios and interactive storytelling. Despite its relatively compact size of 1.54B parameters, the model delivers impressive performance in generating contextually rich dialogues.
Implementation Details
The model utilizes the Transformers architecture and operates in FP16 precision for efficient computation. It's built on the Qwen2-1.5B foundation and supports various generation parameters including temperature, top_p, top_k, and frequency/presence penalties. The implementation includes a sophisticated tokenizer derived from Qwen2.5-1.5B-Instruct, enabling handling of extensive context windows up to 32K tokens.
- Efficient parameter utilization for role-play scenarios
- FP16 precision for optimized performance
- Comprehensive parameter control for generation fine-tuning
- Extended context window support
Core Capabilities
- Specialized role-play dialogue generation
- Dynamic and contextually aware responses
- Support for creative and immersive storytelling
- Efficient resource utilization despite extensive capabilities
- Long-context processing with 32K token input window
Frequently Asked Questions
Q: What makes this model unique?
The model's specialization in role-play scenarios, combined with its efficient architecture and extensive context window, makes it particularly suitable for interactive storytelling applications while maintaining reasonable computational requirements.
Q: What are the recommended use cases?
The model is ideal for interactive fiction, role-playing games, character-based dialogues, and creative writing assistance. Its optimization for role-play scenarios makes it particularly effective in generating contextually appropriate character responses and maintaining consistent narrative flow.