37 stories
Reachy Mini is an open-source robot from Pollen Robotics and Hugging Face, run by a Raspberry Pi with 4 GB of RAM. We wanted it to do what a good host does: greet you by name the second time you walk in, remember what you talked about, and tell you what it saw while you were gone.
A number in a financial report needs its row, column, and unit to mean anything. A chart needs its labels. When you build search over PDFs, that context is part of what you want to retrieve. Perplexity released pplx-embed-v2-late, two late interaction models that search document pages as images.
Build a Semantic Cache for a RAG Pipeline Time: 30 min Level: Intermediate What’s a Semantic Cache Traditional caches operate on an exact match basis, while semantic caches search for the meaning of the key rather than an exact match.
Sparse neural retrieval finally started getting more and more attention (SPARSEUP, MILCO, sparse encoders in Sentence Transformers…). Well, amazing, we like attention!
Classification is one of the oldest problems in machine learning. Unlike modern decoding transformers, most classification models were non-autoregressive by nature.
Google’s new EmbeddingGemma 2 can run on a phone. The model travels light. The vectors need their own suitcase. At 10 million documents, full-size float32 vectors alone take 30.7 GB of RAM.
Monitoring Managed Cloud with Prometheus and Grafana This tutorial will guide you through the process of setting up Prometheus and Grafana to monitor Qdrant databases running in Qdrant Managed Cloud.
As query traffic grows, so does the compute bill for embedding it. On a low-power device, a large model may not fit in memory. A smaller query model could reduce that cost, but switching usually means re-embedding the collection. Constella lets you make that switch.
We rewrote cross-encoder inference in Rust, benchmarked it against Python, and on a small reranking model it came out only 1.1x faster at the median. That is not a sign of a slow Rust implementation.
If you speak more than one language, you know the feeling when the mental switch in your head starts up with the rattling sound of a struggling engine, mixing every word you have to produce into some Denglish, Frenglish, or Spanglish.
We choose embedding models, dimensions, and indexes. The geometry usually comes with the package. But why use a flat space, and what else could we choose? What Is a Manifold, and Where Do Our Vectors Live? A manifold is the space our embeddings live in.
Real world vector search workloads are increasingly large and complex. Enterprises are not using vector search to occasionally search through a couple of PDF files. They are indexing and searching billions of vectors at thousands of requests per second (RPS) and sub 50ms tail latency.
Memory Tiers in Qdrant: What to Use and When A growing vector collection eventually outgrows the RAM it started with: Qdrant handles that by letting you assign dense vectors, the HNSW graph, quantized vectors, payloads, and payload indexes each to whichever memory tier that structure supports,…
GPU-Accelerated HNSW Indexing in Qdrant Time: 45 min Level: Intermediate Output: GitHub Since Qdrant v1.13, Qdrant has supported GPU-accelerated Hierarchical Navigable Small World (HNSW) indexing on self-hosted instances.
A bulk load finishes, and the collection looks ready: every point is in, and the upload call has returned. Then the first queries land, and search takes hundreds of milliseconds, sometimes several seconds at a stretch, while Qdrant’s indexing, merge, and vacuum optimizers work through the backlog…
Hybrid Search in Qdrant A search result can look plausible and still be wrong. Dense retrieval can return a document on the right topic but miss an exact identifier copied into the query. Sparse retrieval can miss a relevant document when the query describes it with terms the corpus doesn’t use.
Your collection works. Queries return in a few milliseconds, results are mostly right, and product keeps forwarding you the ones that aren’t. You open the search API reference and get exact definitions for hnsw_ef, reciprocal rank fusion k, and quantization oversampling.
When Your Collection Outgrows RAM Once a collection no longer fits in RAM, the kernel evicts vector pages, and the next query waits on a disk read to get them back. Quantization buys that memory back.
Showing 18 of 37 stories
Load more storiesGet the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts
