Alkamal01/oribai-14b-hausa-yoruba-v1

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:Apr 13, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

Alkamal01/oribai-14b-hausa-yoruba-v1 is a 7.6 billion parameter instruction-tuned conversational model based on Qwen2.5-14B-Instruct, specifically fine-tuned for Hausa and Yoruba languages. It was trained on 27,498 unique conversational pairs and instruction data, excelling in Yoruba language generation with a perplexity of 3.22. This model is primarily designed for conversational AI applications and factual Q&A in Hausa and Yoruba, offering strong performance in Yoruba and limited but functional performance in Hausa.

Loading preview...

OribAI: A Specialized Hausa & Yoruba Conversational Model

OribAI is an instruction-tuned conversational model developed by Alkamal01, built upon the Qwen2.5-14B-Instruct architecture. With 7.6 billion parameters and a 32768-token context length, it is uniquely fine-tuned for Hausa and Yoruba speakers.

Key Capabilities & Features

  • Bilingual Focus: Optimized for conversational interactions in Hausa and Yoruba.
  • Instruction-Tuned: Trained on 27,498 unique conversational pairs, lexical tasks, and human-annotated instruction data.
  • Strong Yoruba Performance: Achieves a perplexity of 3.22 on Yoruba samples from CohereForAI/aya_dataset, delivering fluent and factually coherent responses.
  • Hausa Factual Q&A: While open-ended Hausa generation can be inconsistent (perplexity 62.54), it performs reliably for short, factual questions.
  • Base Model: Leverages the robust capabilities of Qwen2.5-14B-Instruct.

When to Use This Model

  • Conversational AI in Yoruba: Ideal for chatbots and virtual assistants interacting in Yoruba.
  • Factual Information Retrieval: Suitable for answering direct questions in both Hausa and Yoruba.
  • Language Adaptation: Developers seeking a model specifically trained on African instruction data for these languages.

Limitations to Consider

  • Hausa open-ended generation may exhibit hallucinations or go off-topic.
  • Responses might occasionally mix languages (code-switching).
  • Not evaluated for formal, legal, or medical use cases, and may hallucinate facts, especially for current events.

Finetunes
1 models