hayulalab/Hayula-Rushd-v2-Base

TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Jun 15, 2026License:cc-by-4.0Architecture:Transformer Open Weights Cold

Hayula-Rushd-v2-Base by Hayula Lab is a 7.6 billion parameter language model with a 32768 token context length. This model is specifically designed for next-generation Arabic reasoning tasks. It serves as a foundational base model optimized for complex logical and analytical processing in Arabic.

Loading preview...

Hayula-Rushd-v2-Base: Next-Gen Arabic Reasoning Base

Hayula-Rushd-v2-Base, developed by Hayula Lab, is a 7.6 billion parameter language model engineered for advanced Arabic reasoning. With a substantial context length of 32768 tokens, this model is designed to handle extensive Arabic text inputs, facilitating complex analytical and logical operations.

Key Capabilities

  • Advanced Arabic Reasoning: Optimized as a base model for sophisticated reasoning tasks within the Arabic language.
  • Large Context Window: Supports a 32768 token context, enabling the processing and understanding of long-form Arabic content.

Good For

  • Developing applications requiring deep understanding and logical inference from Arabic text.
  • Research into large language models focused on non-English languages, particularly Arabic.
  • Building specialized Arabic NLP systems that benefit from a robust foundational model.

Licensing

This model is released under the CC-BY-4.0 license, requiring attribution to Hayula Lab when used or adapted.