andre-z
Joined 5 years ago
Karma
210
Submissions
140
Member since
May 11
Recent submissions
Comment · 9 months ago
What do I say? Happy to talk about "fakes". Here is my calendar. Feel free to book a slot. https://qdrant.to/andre-z
Comment · 11 months ago
FAISS is not suitable for production. The dedicated vector search solutions solve all the issues you mentioned: you just store the metadata along with vectors in JSON format. At least, with Qdrant, it works like this: https://qdrant.tech/documentation/concepts/payload/
Comment · 1 year ago
miniCOIL is a contextualized per-word embedding model. It generates extremely small embeddings (8dim or even 4dim) while still preserving the word's context for each word in a sentence.
GitHub https://github.com/qdrant/miniCOIL HuggingFace https://huggingface.co/Qdrant/minicoil-v1
Comment · 1 year ago
"Slowness can arise from a misconfigured index or if filterable attributes aren't listed." ;)
Comment · 1 year ago
Qdrant runs on Linux/Mac/Windows and on x86/ARM processors
Comment · 2 years ago
Any dense embedding model can become a late interaction model.
Comment · 2 years ago
Qdrant lib project https://github.com/tyrchen/qdrant-lib, Qdrant SDK has also support for local mode, which means embeddable https://github.com/qdrant/qdrant-client
Or just use FAISS https://github.com/facebookresearch/faiss
Comment · 2 years ago
The only other Repository is a fork of Qdrant.
Comment · 2 years ago
I can answer how it would be in Qdrant if interested. The index will take around 70GB RAM. New vectors are first placed in a non-indexed segment and are immediately available for search while the index is being built. The vectors and the index can be offloaded to disk. Search will take some milliseconds.