Mxbai Rerank Xsmall is a text ranking model hosted on Hugging Face. Developed by Mixedbread, it belongs to the class of reranker models and is designed to score the relevance of passages to a given query.
The model is based on deberta-v2 and is provided for text-classification tasks. It carries an apache-2.0 license and is distributed in multiple formats that include Transformers, ONNX, Safetensors, and Transformers.js. Users load it directly through the Hugging Face ecosystem for integration into ranking pipelines.
It can be used with the Transformers library by importing AutoTokenizer and AutoModelForSequenceClassification, or through Transformers.js with a pipeline call. The repository also lists support for sentence-transformers and text-embeddings-inference. Deployment options cover local applications, notebooks, and inference providers available on the platform.
The model page indicates 57 likes and is followed under the Mixedbread organization. No pricing information appears because the model is offered as an open-source download under its stated license.
In the Foundation models & chat space, Mxbai Rerank Xsmall takes a focused approach. It focuses on improving the relevance of retrieved documents in RAG systems by reranking initial search results. It is built as an open-source project for developers. The project is open source (Apache-2.0). It ships for the web, the command line, and API, and it can be self-hosted.
It is developed by Mixedbread, and it first shipped in 2023. The project is developed in the open on GitHub with 16.2k stars and 13 commits in the last 90 days. Among its 3 catalogued features are Text Ranking, reranking, and Sequence Classification. It exposes integrations via a public API.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do