hozifa1/Faqih-R1-14B-Islamic-AI

TEXT GENERATIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:14.8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 17, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Faqih-R1-14B-Islamic-AI is a 14.8 billion parameter Qwen2 model developed by hozifa1, fine-tuned from unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit. This model was trained using Unsloth and Huggingface's TRL library, enabling 2x faster training. Its primary differentiation lies in its specialized fine-tuning, making it suitable for applications requiring an Islamic AI context.

Loading preview...

Model Overview

The hozifa1/Faqih-R1-14B-Islamic-AI is a 14.8 billion parameter language model developed by hozifa1. It is based on the Qwen2 architecture and was fine-tuned from the unsloth/DeepSeek-R1-Distill-Qwen-14B-bnb-4bit model. This model leverages the Unsloth library, which facilitated a 2x speedup in its training process, alongside Huggingface's TRL library.

Key Characteristics

  • Base Model: Qwen2 architecture, fine-tuned from DeepSeek-R1-Distill-Qwen-14B.
  • Parameter Count: 14.8 billion parameters.
  • Training Efficiency: Utilizes Unsloth for accelerated training, achieving 2x faster fine-tuning.
  • Context Length: Supports a context window of 32768 tokens.
  • License: Distributed under the Apache-2.0 license.

Potential Use Cases

This model is specifically fine-tuned, suggesting its utility in applications where specialized knowledge or contextual understanding related to its training domain is beneficial. Developers looking for a Qwen2-based model with efficient training origins and a substantial context window may find this model suitable for further adaptation or direct deployment in relevant tasks.