L3.3-MS-Nevoria-70b Open-Source Large Language Model - Free Deployment, Enhance Storytelling and Scene Description Abilities

L3.3 MS Nevoria 70b

Developed by Steelskull

A 70B-parameter large language model based on the Llama 3.3 architecture, enhanced through the merging of multiple superior model components to achieve improved story narration, scene description, and prose detailing capabilities

Large Language Model

Transformers

Open Source License:Other #Low-bias narrative #Multi-character scene control #Long-context handling

Downloads 2,160

Release Time : 1/14/2025

Model Overview

This model combines EVA's storytelling prowess, EURYALE's detailed scene depiction, and Anubis's prose refinement, while reducing positive bias via Negative_LLAMA and incorporating Nemotron traits to create unique weight interaction effects.

Model Features

Enhanced storytelling capability

Incorporates EVA-LLAMA components to deliver superior storytelling and role-playing performance

Detailed scene description

Integrates EURYALE-v2.3 features to generate rich environmental and scene depictions

Reduced positive bias

Mitigates traditional LLM positive bias issues through Negative_LLAMA components

Unique weight interaction

Employs unconventional merging methods to create distinctive weight balancing with special behavioral characteristics

Long-context processing

User-reported capability to handle contexts up to 110,000 tokens without performance degradation

Model Capabilities

Long-form text generation

Role-playing

Story creation

Scene description

Multi-character management

Complex context tracking

Use Cases

Creative writing

Novel writing

Generates lengthy stories with rich details and coherent plots

Users report outputs with exceptional coherence and creativity

Role-playing

Supports complex character interactions and multi-character scenarios

Accurately tracks multiple characters' locations and activities across different settings

Content generation

Detailed scene description

Generates text with immersive environmental details

Provides deeply engaging scene description experiences

🚀 L3.3-MS-Nevoria-70B

L3.3-MS-Nevoria-70B is a powerful model that combines multiple base models, offering excellent storytelling and scene - description capabilities while reducing positive bias.

📚 Documentation

Model Information

Model Name: L3.3-MS-Nevoria-70B
Model Tags:
- L3.3 = Llama 3.3
- MS = Model Stock
- 70B Parameters

Model Composition

Component	Details
EVA-LLAMA-0.1	Storytelling capabilities
EURYALE-v2.3	Detailed scene descriptions
Anubis-v1	Enhanced prose details
Negative_LLAMA	Reduced positive bias
Nemotron-lorablated	Base model

Model Description

This model combines the storytelling capabilities of EVA with the detailed scene descriptions from EURYALE and Anubis. It's further enhanced with Negative_LLAMA to reduce positive bias, with a touch of Nemotron mixed in.

The lorablated model base choice was intentional, creating unique weight interactions similar to the original Astoria model and Astoria V2 model. This "weight twisting" effect, achieved by subtracting the lorablated base model during merging, creates an interesting balance in the model's behavior. While unconventional compared to sequential component application, this approach was chosen for its unique response characteristics.

User Reviews

@Geechan - Discord: "@Steel Have only briefly tested so far, but you really cooked up an amazing merge with this one, and I mean that wholeheartedly. Insane creativity, perfect character adherence and dialogue, loves to slow burn and take its time, minimal sloppy patterns and writing, and such a breath of fresh air in many ways. I'm enjoying my results with 1 temp and 0.99 TFS (close to something like 0.015 min P). Letting the model be creative and wild is so fun and makes me want to RP more. No positivity bias either; violent scenes will result in my death and/or suffering, as they should, and I don't see any soft refusals either. ERP has no skimming of details or refusals like you see on some other L3.3 tunes too"
IGODZOL - Huggingface: "I honestl" (The review seems incomplete here)

📄 License

License Type: other
License Name: eva - llama3.3

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご