maywell/PiVoT-10.7B-Mistral-v0.2-RP
TEXT GENERATIONConcurrent Unit Cost:1Model Size:10.7BQuant:FP8Context Size:4kPublished:Dec 16, 2023License:cc-by-sa-4.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold
PiVoT-10.7B-Mistral-v0.2-RP is a 10.7 billion parameter language model developed by maywell, fine-tuned for roleplay (RP) applications. Based on the PiVoT-10.7B-Mistral-v0.2 architecture, it leverages a private RP dataset collected from ArcaLive Ai Chat Chan. This model features a notable sequence length (Sliding Window) of 8192 tokens, making it suitable for extended conversational contexts in roleplay scenarios.
Loading preview...
Model Overview
PiVoT-10.7B-Mistral-v0.2-RP is a 10.7 billion parameter language model, a specialized variant of the PiVoT-10.7B-Mistral-v0.2 base model. Developed by maywell, its primary distinction lies in its fine-tuning for roleplay (RP) applications.
Key Capabilities
- Roleplay Optimization: The model has been specifically fine-tuned using a private RP dataset sourced from ArcaLive Ai Chat Chan, enhancing its ability to generate coherent and engaging roleplay interactions.
- Extended Context Window: It supports a sequence length (Sliding Window) of 8192 tokens, allowing for longer and more complex conversational histories, which is crucial for immersive roleplay scenarios.
Use Cases
- Interactive Roleplay: Ideal for applications requiring dynamic and context-aware character interactions.
- Creative Storytelling: Can be utilized in generating narrative content with specific character personas and evolving plotlines.