gaocegege

9 ms·

31.
by gaocegege 2y ago
Hi, we’re here to help if you need assistance (via GitHub issue, Discord, or email). Could you let us know the scale of your vectors—are they 1B or 10B?
32.
VectorChord: Store 400k Vectors for $1 in PostgreSQL(blog.pgvecto.rs)
156 points by gaocegege 2y ago|42 comments
33.
by gaocegege 2y ago
Hey everyone! We’ve developed a new PostgreSQL extension that supports 400k vectors for just $1. Check it out!
34.
Should AI workloads run on Kubernetes?(blog.skypilot.co)
7 points by gaocegege 2y ago|1 comments
35.
Pg_bestmatch.rs: Elevate Your PostgreSQL Text Queries with BM25(blog.pgvecto.rs)
3 points by gaocegege 2y ago|0 comments
36.
Unleash the power of sparse vector: Hybrid search using pgvecto.rs and BGE-M3(blog.pgvecto.rs)
1 points by gaocegege 2y ago|0 comments
37.
Pg_bestmatch.rs: Elevate Your PostgreSQL Text Queries with BM25(blog.pgvecto.rs)
4 points by gaocegege 2y ago|0 comments
38.
Pg_bestmatch.rs: Elevate Your PostgreSQL Text Queries with BM25(blog.pgvecto.rs)
3 points by gaocegege 2y ago|0 comments
39.
by gaocegege 3y ago
Gotcha, that makes sense.
40.
by gaocegege 3y ago
Do you mean CUDA API hijacking?
41.
by gaocegege 3y ago
Awesome work! Could you support GPU like modal.com?
42.
by gaocegege 3y ago
In certain situations, we might have to make a trade-off between performance and cost. There are many scenarios where a high level of recall is not necessary.
43.
by gaocegege 3y ago
Hi. Thanks for your comment! I wrote the post to show that adding a reranking step can actually help save memory from the perspective of someone working with databases. In many cases, there may not be a lot of vectors involved. By including
44.
by gaocegege 3y ago
Yeah, it is recall.
45.
Binary vector search is better than FP32 vectors(blog.pgvecto.rs)
137 points by gaocegege 3y ago|29 comments
46.
by gaocegege 3y ago
Thanks for sharing! It is important to note that these outcomes are specific to the openai text-embedding-3-large model
47.
Pgvector vs. Pgvecto.rs: A Comparison for Vector Search in PostgreSQL(blog.pgvecto.rs)
3 points by gaocegege 3y ago|0 comments
48.
Matryoshka Representation Learning (MRL) from the Ground Up(aniketrege.github.io)
2 points by gaocegege 3y ago|0 comments
49.
OpenAI shortened embeddings and adaptive retrieval(docs.pgvecto.rs)
1 points by gaocegege 3y ago|0 comments
50.
Binary Embedding Outperforms OpenAI's Shortened Embedding(docs.pgvecto.rs)
2 points by gaocegege 3y ago|0 comments
51.
Matryoshka Representation Learning (MRL) from the Ground Up(aniketrege.github.io)
3 points by gaocegege 3y ago|0 comments
52.
by gaocegege 3y ago
Do you mean feathr? Welcome PRs!
53.
by gaocegege 3y ago
Yeah, the landscape is using CNCF landscape generator https://github.com/cncf/landscape2 . I also want to introduce the maturity similar to CNCF but not sure how to adopt it. From the maintainer of the project.
54.
by gaocegege 3y ago
Welcome contributions! https://github.com/tensorchord/ai-infra-landscape Open-source contributions can make it better. :-)
55.
by gaocegege 3y ago
Yes, I agree. Here is an example: https://github.com/tensorchord/ai-infra-landscape/pull/5 Huggingface provides model hub and also HF space for inference & deployment. The landscape generator tool support
56.
by gaocegege 3y ago
Thank @histories, I am the maintainer of the landscape. Welcome PRs to add your favorite products! Here is the repo: https://github.com/tensorchord/ai-infra-landscape
57.
What is MRL (OpenAI latest embedding model)?(aniketrege.github.io)
1 points by gaocegege 3y ago|0 comments
58.
Relational Queries and Vector Search(blog.pgvecto.rs)
1 points by gaocegege 3y ago|0 comments
59.
AI Infrastructure Landscape(ai-infra.fun)
1 points by gaocegege 3y ago|0 comments
60.
LLM AI Infrastructure Products Statistics(share.streamlit.io)
2 points by gaocegege 3y ago|0 comments

More ›