Open-weight reranking search results models

14 in the catalogue today. Every one with what it costs, who sells it and where it stands.

Models that take a handful of results a search already found and put them in the right order. They are the cheap second stage of retrieval: an embedding search casts a wide net fast, a reranker reads the candidates properly. Because they only ever see a few documents, they cost little per query, and they usually improve a weak search more than a better embedding model would.

Weights published under a licence that lets you run them where you like and sell what you build — Apache 2.0, MIT and their kin impose little beyond keeping the notice. This is the list to start from if you need the model on your own hardware, in your own region, or simply want no third party between you and it. The trade is that hosting is now your problem, and the prices shown beside each one are what somebody else charges to do it for you.

• Granite Embedding 278M MultilingualIBM watsonx.ai$0.11per Mtok in1 selling
• Qwen3 8B (embeddings)Alibaba3rd of 238· 3 boards$0.01–$0.1per Mtok in6 selling
• Qwen3-Embedding-0.6BAlibaba25th of 183· 2 boards$0.01–$0.04per Mtok in4 selling
• Qwen3-Embedding-4BAlibaba5th of 238· 3 boards$0.01–$0.02per Mtok in4 selling
• Qwen3-Reranker-0.6BAlibaba$0.01per Mtok in1 selling
• Qwen3-Reranker-4BAlibaba$0.025per Mtok in1 selling
• Qwen3-Reranker-8BAlibaba$0.2per Mtok in2 selling
• Qwen3-VL-Embedding-2BAlibaba
• Qwen3-VL-Embedding-8BAlibaba
• bce-embedding-base_v1maidalun1020
• bge-reranker-largeBAAI
• bge-reranker-v2-m3BAAI$0.01per Mtok in2 selling
• gte-basethenlper$0.005per Mtok in2 selling
• gte-largethenlper$0.01–$0.09per Mtok in3 selling

Wider