Shinzmann/naija-petro-8b

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Mar 20, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

Naija-Petro 8B is an 8 billion parameter domain-specific large language model developed by Shinzmann, fine-tuned on Qwen3-8B for petroleum engineering tasks. It is optimized for fast inference and deployment, covering topics such as drilling, reservoir, production, and enhanced oil recovery. This model utilizes QLoRA fine-tuning with Unsloth on over 20,000 synthetic instruction-response pairs. It is designed for specialized applications in the petroleum industry, offering a lightweight solution for domain-specific queries.

Loading preview...

Naija-Petro 8B: Domain-Specific Petroleum Engineering LLM

Naija-Petro 8B is a specialized 8 billion parameter language model, fine-tuned by Shinzmann from the Qwen3-8B architecture. It is specifically designed for applications within the petroleum engineering domain, offering a lightweight solution for fast inference and deployment.

Key Capabilities & Features

  • Domain Expertise: Covers core petroleum engineering areas including drilling, reservoir management, production, completions, enhanced oil recovery (EOR), and well testing.
  • Efficient Fine-tuning: Utilizes QLoRA 4-bit fine-tuning with Unsloth, resulting in 2x faster training and 70% less VRAM consumption compared to standard methods.
  • Training Data: Fine-tuned on over 20,000 synthetic instruction-response pairs tailored to petroleum engineering concepts.
  • Accessibility: Designed for easy deployment, including on free HuggingFace ZeroGPU Spaces, and available for use with Ollama.

Good For

  • Petroleum Engineering Applications: Ideal for developers and engineers needing a specialized LLM for tasks within the oil and gas industry.
  • Resource-Constrained Environments: Its lightweight nature and optimization for fast inference make it suitable for environments with limited computational resources.
  • Rapid Prototyping: Provides a quick-start option for integrating domain-specific AI capabilities into petroleum engineering workflows.

Limitations

  • Outputs should always be validated by qualified engineers before operational use.
  • The model is English-only and not intended for general chat or non-domain-specific conversations.

Finetunes
2 models