back

SCB 10X · 8B · Dense

Thai/English bilingual chat model built on Llama 3.1

0 downloads 0 likes 2024-12 88K context

Use Cases

chat multilingual code

Quantization Options

Quant Bits VRAM Quality Status
Q2_K 2 3.1 GB low
Q3_K_M 3 4.1 GB moderate
Q4_K_M 4 4.6 GB good
Q5_K_M 5 5.6 GB good
Q6_K 6 6.6 GB excellent
Q8_0 8 8.7 GB excellent
F16 16 16.9 GB lossless

About this model

Llama3.1-Typhoon2-8B: Thai Large Language Model (Instruct) - Q4_K_M quantized

Llama3.1-Typhoon2-8B-instruct is a instruct Thai 🇹🇭 large language model with 8 billion parameters, and it is based on Llama3.1-8B.

By using this model, you agree to the OpenTyphoon Terms and Conditions and acknowledge the Privacy Notice: https://opentyphoon.ai/tac · https://opentyphoon.ai/privacy

*To acknowledge Meta’s effort in creating the foundation model and to comply with the license, we explicitly include “llama-3.1” in the model name.

Run

ollama run scb10x/llama3.1-typhoon2-8b-instruct

Performance

Instruction-Following & Function Call Performance

Typhoon2 Llama 8B General Performance

Specific Domain Performance (Math & Coding)

TTyphoon2 Llama 8B Specific Domain Performance

Long Context Performance

Typhoon2 Llama 8B Long Context Performance

Detail Performance

Model IFEval - TH IFEval - EN MT-Bench TH MT-Bench EN Thai Code-Switching(t=0.7) Thai Code-Switching(t=1.0) FunctionCall-TH FunctionCall-EN GSM8K-TH GSM8K-EN MATH-TH MATH-EN HumanEval-TH HumanEval-EN MBPP-TH MBPP-EN
Llama3.1 8B Instruct 58.04% 77.64% 5.109 8.118 93% 11.2% 36.92% 66.06% 45.18% 62.4% 24.42% 48% 51.8% 67.7% 64.6% 66.9%
Typhoon2 Llama3 8B Instruct 72.60% 76.43% 5.7417 7.584 98.8% 98% 75.12% 79.08% 71.72% 81.0% 38.48% 49.04% 58.5% 68.9% 60.8% 63.0%

Link

Arxiv Huggingface