Which model would you like to support?
HuggingFace Link: https://huggingface.co/ibm-granite/granite-embedding-small-english-r2
ONNX Weights Link: https://huggingface.co/onnx-community/granite-embedding-small-english-r2-ONNX
What are the main advantages of this model?
It is a highly popular, lightweight embedding model that performs exceptionally well on the MTEB leaderboard for its tiny size (under 50M parameters). It's a great fit for FastEmbed's goal of fast, serverless embeddings.
Note: I have already implemented this, added it to the builtin sentence embeddings registry, and verified the canonical test values locally. I will be opening a Pull Request to address this immediately!
Which model would you like to support?
HuggingFace Link: https://huggingface.co/ibm-granite/granite-embedding-small-english-r2
ONNX Weights Link: https://huggingface.co/onnx-community/granite-embedding-small-english-r2-ONNX
What are the main advantages of this model?
It is a highly popular, lightweight embedding model that performs exceptionally well on the MTEB leaderboard for its tiny size (under 50M parameters). It's a great fit for FastEmbed's goal of fast, serverless embeddings.
Note: I have already implemented this, added it to the builtin sentence embeddings registry, and verified the canonical test values locally. I will be opening a Pull Request to address this immediately!