Do you know anyone working on a large text completion model based on it?
Do you have/plan to have a text embeddings model?
Yes, we are training text embedding models right now. And also have plans to open-source some of them!
In addition, we train encoders for different modalities with retrieval purposes. For example, video data.