mxbai-rerank-xsmall-v1

mixedbread-ai

mixedbread-ai/mxbai-rerank-xsmall-v1

RetrievalOpen weightsApache 2.0Verified metrics

View on Hugging Face ↗

About

Very small reranker for latency-constrained pipelines where a large cross-encoder is too slow.

Metrics

Downloads

665,927

last 30 days

Hub likes

59

Parameters

71M

published total

How it compares in Retrieval

Sizemxbai-rerank-xsmall-v1 sits in <1B — larger than 15% of the 33 models in this category with a published size.

<1B30

1-10B3

Details

Model ID: mixedbread-ai/mxbai-rerank-xsmall-v1
Developer: mixedbread-ai
Category: Retrieval
Access: Open weights
License: Apache 2.0
Source: Hugging Face
Data provenance: Metrics fetched per-model from the Hugging Face API.
Link: https://huggingface.co/mixedbread-ai/mxbai-rerank-xsmall-v1

More from mixedbread-ai

mxbai-embed-large-v1mixedbread-ai/mxbai-embed-large-v1Embedding mxbai-embed-2d-large-v1mixedbread-ai/mxbai-embed-2d-large-v1Embedding deepset-mxbai-embed-de-large-v1mixedbread-ai/deepset-mxbai-embed-de-large-v1Embedding

Other Retrieval models

bge-reranker-largeBAAI/bge-reranker-largeRetrieval splade-cocondenser-selfdistilnaver/splade-cocondenser-selfdistilRetrieval splade-cocondenser-ensembledistilnaver/splade-cocondenser-ensembledistilRetrieval ms-marco-MiniLM-L6-v2cross-encoder/ms-marco-MiniLM-L6-v2Retrieval ms-marco-MiniLM-L4-v2cross-encoder/ms-marco-MiniLM-L4-v2Retrieval gte-reranker-modernbert-baseAlibaba-NLP/gte-reranker-modernbert-baseRetrieval