migtissera/Tess-70B-v1.6

TEXT GENERATIONConcurrency Cost:4Model Size:69BQuant:FP8Ctx Length:32kPublished:Mar 2, 2024License:llama2Architecture:Transformer0.0K Open Weights Cold

Tess-70B-v1.6 is a 69 billion parameter general-purpose large language model developed by migtissera, built upon the Miqu/LLaMA-2-70B base architecture. This model is designed for broad applicability across various natural language processing tasks, offering a substantial context length of 32768 tokens. It aims to provide a versatile foundation for developers seeking a robust and adaptable LLM solution.

Loading preview...

Overview

Tess-70B-v1.6, also known as Tesoro (Italian for "Treasure"), is a 69 billion parameter general-purpose large language model developed by migtissera. It is built on the Miqu/LLaMA-2-70B base, indicating a strong foundation for diverse applications. With a context length of 32768 tokens, it can process and generate longer sequences of text, making it suitable for complex tasks requiring extensive context understanding.

Key Characteristics

  • Model Size: 69 billion parameters, offering significant capacity for understanding and generation.
  • Base Architecture: Derived from Miqu/LLaMA-2-70B, leveraging established and robust LLM design principles.
  • Context Window: Supports a 32768-token context length, beneficial for detailed conversations, document analysis, and multi-turn interactions.

Prompt Format

The model utilizes a specific prompt format for optimal interaction:

SYSTEM: <ANY SYSTEM CONTEXT>
USER: 
ASSISTANT:

This structure allows for clear separation of system instructions and user input, guiding the model's responses effectively.

Good For

  • General-purpose text generation and understanding.
  • Applications requiring a large context window.
  • Developers looking for a robust LLaMA-2-based model with a substantial parameter count.