Overview

Sailor2 is a community-driven initiative that brings cutting-edge multilingual language models to South-East Asia (SEA). It is designed to address the growing demand for diverse, robust, and accessible language technologies in the region. Built upon the foundation of Qwen 2.5, Sailor2 is continuously pre-trained on 500B tokens, significantly improving its support for 15 languages with a unified model. These languages include English, Chinese, Burmese, Cebuano, Ilocano, Indonesian, Javanese, Khmer, Lao, Malay, Sundanese, Tagalog, Thai, Vietnamese, and Waray.

Sailor2 is available in three sizes: 1B, 8B, and 20B, which are expansions from the Qwen2.5 base models of 0.5B, 7B, and 14B, respectively. These models serve a wide range of applications, from production use to research and speculative decoding, ensuring accessibility to advanced language technologies across SEA.

Variants

No Variant Cortex CLI command
1 gguf cortex run sailor-2

Use it with Jan (UI)

  1. Install Jan using Quickstart
  2. Use in Jan model Hub:
    cortexhub/sailor-2
    

Use it with Cortex (CLI)

  1. Install Cortex using Quickstart
  2. Run the model with command:
    cortex run sailor-2
    

Credits

  • Author: Community-driven (Sailor2 Initiative)
  • Converter: [Homebrew]
  • Original License: [Licence]
  • Papers: [Paper]
Downloads last month
666
GGUF
Model size
988M params
Architecture
qwen2

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.