new5558
commented
12 months ago Result: https://colab.research.google.com/drive/1B1yiRkEdtzEfXGu-MJqRLpWZgsOBKdFc?usp=sharing
Most Interesting Model now: QWen-7B
- Thai Tokenizer
- Trained for 2 Trillion tokens mainly in English and Chinese
- Multimodal Support
- Comparable performance with SOTA OSS LLM
- Commercial use available for < 100M users
Expecting Output