SteelStorage/L3.3-MS-Evalebis-70b
SteelStorage/L3.3-MS-Evalebis-70b is a 70 billion parameter language model developed by SteelSkull, built upon the Llama 3.3 architecture. This model is a merge of EVA-LLAMA-0.1, EURYALE-v2.3, and Anubis-v1, designed to combine their strengths in robust storytelling, prose, and detailed scene descriptions. With a context length of 32768 tokens, it is optimized for creative writing tasks requiring rich narrative and descriptive output.
Loading preview...
Overview
SteelStorage/L3.3-MS-Evalebis-70b is a 70 billion parameter language model created by SteelSkull, based on the Llama 3.3 architecture. It is a merged model, combining the characteristics of EVA-LLAMA-0.1, EURYALE-v2.3, and Anubis-v1. The primary goal of this merge was to integrate the robust storytelling capabilities of EVA, the detailed prose and scene descriptions from EURYALE, and the narrative strengths of Anubis, while preserving the positive attributes of each base model.
Key Capabilities
- Enhanced Storytelling: Designed to produce rich and engaging narratives by leveraging the strengths of its constituent models.
- Detailed Prose: Excels at generating descriptive and nuanced text, ideal for creative writing and immersive content.
- Scene Description: Capable of creating vivid and intricate scene details, contributing to a more immersive user experience.
- Llama 3.3 Foundation: Benefits from the underlying architecture of Llama 3.3, providing a strong base for language generation.
Recommended Usage
This model is particularly well-suited for applications requiring high-quality creative writing, detailed narrative generation, and rich descriptive content. Users can explore recommended templates and system prompts like "LLam@ception" and "DWK-SP0.02" for optimal performance, which are linked in the original README. Quantized versions (GGUF) are also available from community contributors like bartowski and mradermacher for more efficient deployment.