mohdAlal1/Nafha-Llama3.1-8B-Perfumery-Expert-v1

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Apr 5, 2026License:llama3.1Architecture:Transformer0.0K Featherless Exclusive Cold

Nafha-Llama3.1-8B-Perfumery-Expert-v1 by mohdAlal1 is an 8.03 billion parameter LLaMA 3.1 model fine-tuned using QLoRA for expert-level perfumery knowledge. This model excels as a knowledgeable perfumery consultant and fragrance formulation assistant, trained on a custom dataset derived from 35 perfumery reference books. It specializes in ingredient knowledge, accord building, fragrance pyramid classification, and generating structured perfume formulas, making it ideal for specialized perfumery applications.

Loading preview...

Nafha-Expert: Llama 3.1 8B Perfumery Specialist

Nafha-Expert is a specialized large language model, fine-tuned by mohdAlal1, designed to act as an expert perfumery consultant and fragrance formulation assistant. Built on the Meta Llama 3.1 8B Instruct base model and fine-tuned using QLoRA, this model is the AI core of the NeuroScent project, aiming to deliver a personalized perfume creation experience.

Key Capabilities

  • Expert Perfumery Knowledge: Deep understanding of aroma chemicals, natural ingredients, isolates, and their olfactory profiles.
  • Accord Building & Formulation: Guidance on constructing balanced fragrance accords and understanding fragrance pyramid structures.
  • NeuroComposer Integration: Translates psychological and emotional profiles into structured perfume formulas, compatible with NeuroScent's dispensing model.
  • Specialized Q&A: Trained on approximately 19,200 instruction-response pairs from 35 perfumery reference books and scientific texts.

Good For

  • Expert perfumery Q&A and consultation.
  • Generating fragrance formulation guidance and accord building advice.
  • Educational resources for olfaction and aroma chemistry.
  • Integration into specialized systems requiring deep perfumery domain expertise, such as the NeuroScent personalized perfume system.

This model is domain-specific and performs best with a detailed system prompt to maintain its expert persona, as demonstrated in the provided inference examples.