Shinzmann/naija-petro

TEXT GENERATIONPricing:Input $0.408 / Cached $0.0816 / Output $1.972Concurrent Unit Cost:2Model Size:32BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Mar 19, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Shinzmann/naija-petro is a 32 billion parameter Qwen3-based instruction-tuned causal language model developed by the Naija-Petro project, fine-tuned on approximately 20,000 synthetic petroleum-engineering instruction-response pairs. This model excels at technical question answering and explanation across various petroleum-engineering subdomains, serving as a study aid and engineering decision-support tool. It is specifically optimized for domain-specific knowledge in petroleum engineering, offering detailed, technically accurate answers with equations and units.

Loading preview...

Overview

Shinzmann/naija-petro is a 32 billion parameter instruction-tuned model, built upon the Qwen3-32B architecture. Developed by the Naija-Petro project, it has been fine-tuned using QLoRA on approximately 20,000 synthetic petroleum-engineering instruction-response pairs. This model represents the highest-quality variant within the Naija-Petro family, offering deep domain-specific knowledge in petroleum engineering.

Key Capabilities

  • Technical Question Answering: Provides precise and technically accurate answers across petroleum-engineering subdomains, including concepts, equations, derivations, workflow guidance, and terminology.
  • Domain Expertise: Covers drilling, reservoir, production, completions, Enhanced Oil Recovery (EOR), well testing, and petroleum geoscience.
  • Decision Support: Functions as a study aid and engineering decision-support tool.
  • Foundation for RAG Systems: Can serve as a backbone for retrieval-augmented assistants, further domain fine-tuning, or distillation into smaller models.

Good For

  • Petroleum Engineering Professionals: Seeking detailed explanations and answers to complex technical questions.
  • Students and Researchers: Requiring a specialized AI assistant for petroleum-related studies and research.
  • Developing Specialized AI Tools: As a base model for building more advanced, domain-specific AI applications, especially when paired with a RAG system for up-to-date and localized information (e.g., Nigeria-specific regulations).

Note: While highly specialized, the model's knowledge is static as of training. For Nigeria-specific facts or current regulations, it is recommended to pair it with the Naija-Petro RAG system to ground answers in verifiable sources. Outputs should always be validated by qualified engineers and primary sources.

Finetunes
4 models