sethuiyer/Dr.Samantha-8B

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kLicense:otherArchitecture:Transformer0.0K Cold

Dr. Samantha-8B by sethuiyer is an 8 billion parameter Llama-3-based model, merged using DARE TIES, specifically designed as a virtual medical assistant. It integrates specialized medical knowledge from models like medllama3-v20 and Medichat-V2-Llama3-8B with conversational capabilities, providing empathetic, knowledgeable, and thorough support for health-related inquiries. With an 8192 token context length, it excels at offering reliable medical advice and differential diagnoses.

Loading preview...

Dr. Samantha-8B: An Empathetic Medical AI Assistant

Dr. Samantha-8B is an 8 billion parameter large language model developed by sethuiyer, built upon the NousResearch/Meta-Llama-3-8B base. This model is uniquely engineered as a virtual medical assistant, combining deep medical expertise with empathetic patient interaction.

Key Capabilities

  • Specialized Medical Knowledge: Integrates knowledge from dedicated medical LLMs such as ProbeMedicalYonseiMAILab/medllama3-v20 and sethuiyer/Medichat-V2-Llama3-8B.
  • Empathetic Interaction: Blends conversational capabilities from models like ruslandev/llama-3-8b-samantha and NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS to ensure approachable and compassionate patient care.
  • Differential Diagnosis: Capable of providing potential differential diagnoses based on described symptoms, as demonstrated in its usage examples.
  • Reliable Health Advice: Designed to offer professional, knowledgeable, and thorough support for a wide range of health-related inquiries.

Unique Approach

Dr. Samantha-8B was created using the DARE TIES merge method, a technique that combines multiple specialized models to leverage their individual strengths. This blend ensures the model is both medically informed and capable of natural, human-like conversation, making it a trustworthy companion for health advice.

Good for

  • Applications requiring a virtual medical assistant.
  • Generating empathetic and knowledgeable responses to health questions.
  • Providing preliminary differential diagnoses based on user-provided symptoms.
  • Integrating AI for patient support and information in healthcare contexts.

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