Nekochu/Confluence-Renegade-7B
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Mar 19, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Nekochu/Confluence-Renegade-7B is a 7 billion parameter language model developed by Nekochu, built using a DARE linear and Slerp merge of several roleplay-focused models, including BuRP_7B and Mistral-7B-v0.1 as a base. Designed for creative writing and roleplay scenarios, it aims to provide unconstrained and diverse narrative generation. The model is specifically optimized for generating engaging and unrestricted conversational content, making it suitable for applications requiring flexible and imaginative text outputs.

Loading preview...

Confluence-Renegade-7B Overview

Confluence-Renegade-7B is a 7 billion parameter language model created by Nekochu, specifically engineered for roleplay (RP) and creative writing tasks. This model is a unique merge, combining several specialized RP models using a DARE linear and Slerp merge method, with Mistral-7B-v0.1 serving as its base architecture. The name "Confluence" signifies the merging of diverse RP models, while "Renegade" highlights its focus on unconstrained and no-guardrail content generation.

Key Capabilities

  • Specialized Roleplay Generation: Designed to excel in generating dynamic and engaging roleplay narratives.
  • Unrestricted Content: Aims to provide outputs without typical content filters, catering to diverse creative scenarios.
  • Mergekit Architecture: Built using mergekit with a combination of dare_linear and slerp methods, integrating models like Erosumika-7B, Infinitely-Laydiculous-7B, Kunocchini-7b-128k-test, EndlessRP-v3-7B, daybreak-kunoichi-2dpo-7b, and BuRP_7B.

Good For

  • Creative Writing: Ideal for generating imaginative stories, character dialogues, and complex narrative arcs.
  • Roleplay Applications: Suited for interactive fiction, chatbot roleplay, and scenarios requiring flexible, uninhibited responses.
  • Experimental Content Generation: Useful for developers and users exploring the boundaries of LLM outputs in creative contexts.