Model2Vec
Models
Our collection of static embedding models
Model List
We provide a number of models that can be used out of the box. These models are available on the HuggingFace hub and can be loaded using the from_pretrained
method. The models are listed below:
Model | Language | Sentence Transformer | Params | Task |
---|---|---|---|---|
potion-base-32M | English | bge-base-en-v1.5 | 32.3M | General |
potion-multilingual-128M | Multilingual | bge-m3 | 128M | General |
potion-retrieval-32M | English | bge-base-en-v1.5 | 32.3M | Retrieval |
potion-base-8M | English | bge-base-en-v1.5 | 7.5M | General |
potion-base-4M | English | bge-base-en-v1.5 | 3.7M | General |
potion-base-2M | English | bge-base-en-v1.5 | 1.8M | General |
Recommended Models
Which model is best for you depends on your use case. We recommend the following models:
- General Use (English): potion-base-32M is the most performant model for general English tasks.
- General Use (Multilingual): potion-multilingual-128M is the best choice for multilingual tasks.
- Retrieval (English): potion-retrieval-32M is optimized for English retrieval tasks and is the best choice for this use case.
- Lightweight (English): potion-base-8M is the best choice for lightweight applications since it’s only ~30mb on disk.
Note that all our models are the same speed, so you can choose the model based on your size and performance requirements.