I made a way to migrate between embedding models without re-embedding your entire corpus

So I was playingw ith embedding models I saw that when you upgrade from model A to B, you face a very big backfilling cost Ie, suppose you have a 1b vectors from model A, and then you want to use model B. This would mean you have to re-embed all of your documents with model B before you can even serve with the model, and on an H100, it would take ~108 days (qwen embed 8b, 106 docs/second). But I found an easier way to do it. The method is really simple; from the old index made with the source model, take K documents and rerank them with the new model. We see that when K is sufficient, the retrieval quality is the same as target model. (determining k is the hard part). I've tested 63 migrations on upto 1 million documents. The best result I got was upgrading qwen4b -> to 8b, and at 50 documents, it was the same as native retrieval. This method forgos the expensive upfront re-embedding cost, as you can take documents straight from the old index. embedflow works with qdrant, pgvector, faiss, and can be easily downloaded with pypi pip install embedflow the github is public: https://github.com/arnsri33/embedflow I want you guys to try it out, and see if you guys can use it in your own workflow.