BGE embedding is a general Embedding Model. It is pre-trained using retromae and trained on large-scale pair data using contrastive learning. Note that the goal of pre-training is to reconstruct the text, and the pre-trained model cannot be used for similarity calculation directly, it needs to be fine-tuned
https://api.deepinfra.com/v1/openai/v1/embeddings| Model | Context | Max out |
|---|---|---|
| bge-base-en-v1.5 | — | — |
| bge-en-icl | — | — |
| bge-large-en-v1.5 | — | — |
| bge-m3-multi | — | — |
| bge-m3 | — | — |
/v1/models/deepinfra/BAAI/bge-base-en-v1.5