BruhzWater/Edens-Fall-L3.3-70b-0.3c

Hugging Face
TEXT GENERATIONConcurrency Cost:4Model Size:70BQuant:FP8Ctx Length:32kArchitecture:Transformer0.0K Warm

BruhzWater's Edens-Fall-L3.3-70b-0.3c is a 70 billion parameter language model, part of the Eden-L3.3 series, specifically designed for creative writing and roleplay with a 32K context length. This model is the final stage of a three-part Multi-SLERP merge, aiming to produce lengthy and non-repetitive narrative responses. It integrates components from various specialized models, including those focused on detailed descriptions, creative prose, and roleplay, to enhance its storytelling capabilities.

Loading preview...

BruhzWater/Edens-Fall-L3.3-70b-0.3c: A Specialized Model for Creative Writing

BruhzWater's Edens-Fall-L3.3-70b-0.3c is a 70 billion parameter language model, developed as the final stage of a three-part merge using the Multi-SLERP method. Its primary objective is to excel in storytelling and roleplay, focusing on generating lengthy and diverse responses without repeating narrative beats. The model is built upon a foundation of several specialized models, each contributing unique strengths to its overall performance.

Key Capabilities

  • Enhanced Creative Writing: Designed to produce detailed and engaging narratives.
  • Roleplay Optimization: Specifically fine-tuned for interactive roleplaying scenarios.
  • Non-Repetitive Output: Aims to avoid common LLM pitfalls of repetitive story elements.
  • Extensive Context Window: Supports a 32,768-token context length, beneficial for longer interactions and complex plots.
  • Multi-SLERP Merge: Combines diverse models like Serpents-Tongue, Apocrypha, and Liliths-Whisper, integrating components optimized for detail, creative prose, and roleplay.

Good For

  • Creative Writers: Ideal for generating story drafts, character dialogues, and plot developments.
  • Roleplay Enthusiasts: Provides a robust engine for immersive and dynamic roleplaying experiences.
  • Applications Requiring Long-Form Narrative: Suitable for tasks where extended, coherent, and varied text generation is crucial.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p