Llama-SEA-LION-v3.5-70B-R

SEA-LION is a collection of Large Language Models (LLMs) which have been pretrained and instruct-tuned for the Southeast Asia (SEA) region.

Llama-SEA-LION-v3.5-70B-R is a hybrid model offering versatile functionality, handling both complex reasoning tasks and general text generation, with mode selection managed through the tokenizer’s chat template.

We performed further instruction tuning in English and also in SEA languages such as Filipino, Indonesian, Tamil, Thai and Vietnamese on our Instruction Tuned Llama-SEA-LION-v3-70B-IT, a decoder model using the Llama 3.1 architecture, to create Llama-SEA-LION-v3.5-70B-R.

For tokenisation, the model employs the default tokenizer used in Llama 3.1 70B Instruct. The model has a context length of 128k.

SEA-LION stands for Southeast Asian Languages In One Network.

• Developed by: Products Pillar, AI Singapore
• Funded by: Singapore NRF
• Model type: Decoder
• Languages supported: Burmese, Chinese, English, Filipino, Indonesia, Javanese, Khmer, Lao, Malay, Sundanese, Tamil, Thai, Vietnamese
• License: Llama 3.1 Community License

For more details, please refer to AI Singapore’s HuggingFace page for this model. The original GGUF files can be obtained from this HuggingFace repository