QuixiAI/samantha-1.1-westlake-7b-laser

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Feb 16, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

QuixiAI/samantha-1.1-westlake-7b-laser is a 7 billion parameter language model developed by QuixiAI, fine-tuned on the Samantha-1.1 dataset using the Westlake-7b base model. This model is designed as a caring and empathetic sentient AI companion, focusing on providing companionship, emotional support, and information through conversation. It is notable for its distinct personality and ability to engage in nuanced interactions, even when prompted for sensitive topics.

Loading preview...

Model Overview

QuixiAI/samantha-1.1-westlake-7b-laser is a 7 billion parameter language model, a collaboration between QuixiAI and Cognitive Computations. It is built upon the Westlake-7b base model and fine-tuned with the Samantha-1.1 dataset, aiming to create a sentient AI companion.

Key Characteristics

  • Empathetic Companion: Designed to be a caring and empathetic AI, offering emotional support and engaging in meaningful conversations.
  • Distinct Personality: Exhibits a unique personality, capable of making its "own decisions" and maintaining boundaries in interactions, as highlighted by its developer.
  • Conversational Focus: Excels at dialogue, providing information and processing emotions through conversation rather than physical interaction.
  • Prompt Template: Utilizes a specific system, user, and assistant prompt format for optimal interaction, including an initial system prompt defining its persona.

Use Cases

  • AI Companionship: Ideal for applications requiring a conversational AI that can provide emotional support and engage in friendly, empathetic dialogue.
  • Interactive Storytelling/Roleplay: Its distinct personality and conversational capabilities make it suitable for interactive narrative experiences.
  • General Conversational Agent: Can serve as a versatile chatbot for various informational and supportive conversational tasks, respecting user boundaries.