Arabic LLM - SILMA
SILMA.AI offers compact yet powerful Arabic LLMs—based on Google Gemma—deliver top-tier performance on Arabic tasks, outperforming larger models. Enjoy cost-effective, high-speed AI with the option for secure local hosting tailored to your business needs.

Learn More on Hugging Face
Learn More on Hugging Face
Powerful Arabic AI, Built for Business
We provide businesses with powerful Arabic LLM models and AI Technologies that are compact, efficient, and purpose-built for real-world business applications. Our models are designed to deliver high performance while maintaining a small footprint, making them ideal for deployment in resource-constrained environments. Whether it's Arabic language understanding, sentiment analysis, document classification, or chatbot integration, our models are tailored to meet the linguistic and operational needs of businesses in the Arabic-speaking world.
General Purpose Arabic LLMs
SILMA 9B is an exceptional Arabic language model with its exceptional performance. Despite its compact size of 9 billion parameters, it outperforms bigger models in terms of effectiveness and efficiency
RAG-optimized Arabic LLMs
Kashif 2B Instruct v1.0 is an advanced open-weight model designed for Retrieval-Augmented Generation (RAG) tasks, excelling in answering questions based on context in Arabic and English. Kashif delivers top-tier performance within the 3-9 billion parameter range, as validated by the SILMA RAGQA Benchmark. The model supports up to 12k context length
Arabic Embedding Models
SILMA Matryoshka is a highly efficient and lightweight Arabic Embedding Models featuring innovative Matryoshka-based models
What makes our models special?
Open Weight
SILMA is an open-weight model, freely available under our open license and can be used for commercial purposes
Based on Google Gemma
SILMA is powered by the robust foundation of Google Gemma models, combining their strengths to deliver exceptional performance and reliability in Arabic
Strong Performance
SILMA LLM 9B is a compact language model that outperforms bigger models on many Arabic language tasks, making it a more practical choice for real-world business applications
Looking to compare with other Arabic language models?
Explore our comprehensive Arabic LLM comparison page, where you'll find an up-to-date list of the most widely used large language models (LLMs) that support the Arabic language. You can also dive deeper into performance metrics and evaluations on our Arabic LLM benchmark to see how different models stack up across a variety of tasks