Indexnusrefather/Palette-RP-9B-2609-v0.05
VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 29, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold
Indexnusrefather/Palette-RP-9B-2609-v0.05 is a 9 billion parameter language model, fine-tuned from Qwen 3.5 9B. This model is specifically optimized for roleplay scenarios, particularly ERP, by training on Hy4 RP data with reasoning capabilities intentionally disabled. It is designed to excel in creative and interactive narrative generation within roleplaying contexts.
Loading preview...
Palette-RP-9B-2609-v0.05: Roleplay Focused Qwen 3.5 Finetune
This model, developed by Indexnusrefather, is an experimental 9 billion parameter language model based on Qwen 3.5 9B. It has been specifically fine-tuned for enhanced performance in roleplaying (RP) scenarios, with a particular focus on ERP (Erotic Roleplay).
Key Characteristics & Training:
- Base Model: Qwen 3.5 9B.
- Parameter Count: 9 billion.
- Context Length: 32768 tokens.
- Training Data: Utilizes Hy4 RP data.
- Reasoning Disabled: The model's reasoning capabilities were intentionally disabled during training to prioritize creative and narrative generation over logical inference.
- Development: Trained on a 5070 Ti, indicating a focus on accessible, home-based fine-tuning.
Primary Use Case:
- Roleplay Generation: Optimized for generating engaging and coherent responses within roleplaying contexts, especially those requiring detailed and immersive narrative contributions.
- ERP Focus: Designed to be particularly effective for Erotic Roleplay scenarios due to its specialized training data and disabled reasoning, which can lead to more uninhibited creative output.