tencent

/

R3-embedding-0.6b

like 27

Follow

Tencent 11.7k

Hello @xiaoxiaoshadiao and team,

First of all, congratulations on the release! I'm a big fan of using the retrieve & rerank pipeline for finding the correct skills to use, so it's nice to see more progress here.
I wanted to make it even easier for users to understand how to use it, so I updated the README here. It also removes trust_remote_code=True, as that's actually not even necessary!

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("tencent/R3-embedding-0.6b")
query_embedding = model.encode_query("I need to compose music")
document_embeddings = model.encode_document([  # The format is "name | description | skill_md"
    "music-composer | Composes original music | Creates music for various media formats ...",
    "music-lyricist | Writes lyrics for songs | Creates lyrics for various music genres ...",
    "music-editor | Edits and mixes music tracks | Provides audio editing and mixing services ...",
])
similarities = model.similarity(query_embedding, document_embeddings)
print(similarities)
# tensor([[0.7410, 0.5510, 0.5028]])

With the output similarity it's immediately clear that it works as expected.

• Tom Aarsen