Category product updates
Photo by Shiro hatori on Unsplash
High performance feeding with Vespa CLI
Vespa CLI can now feed large sets of documents to Vespa efficiently.
Photo by Taylor Vick on Unsplash
Private regional endpoints in Vespa Cloud
Set up private endpoint services on your Vespa Cloud application, and access them from your own VPC, in the same region, through the cloud provider's private network.
Photo by Sandro Katalina on Unsplash
GPU-accelerated ML inference in Vespa Cloud
Today we're introducing support for GPU-accelerated ONNX model inference in Vespa, together with support for GPU instances in Vespa Cloud!
Basic HTTP testing of Vespa applications
Write basic tests for a Vespa application in JSON format, run them with the Vespa CLI, and use them to verify production deployments in Vespa Cloud.
Photo by Ilya Pavlov on Unsplash
Vespa Product Updates, September 2021
Advances in Vespa features and performance include Vespa CLI, nearest neighbor performance improvements, mTLS security, improved write throughput and Sentencepiece Encoder.
Introducing Vespa CLI
The official command-line tool for Vespa is now available.
Photo by Scott Graham on Unsplash
Vespa Product Updates, July 2021
Advances in Vespa features and performance include HTTP/2 feed and query endpoints, ONNX Runtime integration for stateless model inference, and the new open Vespa Factory build and test system.
Photo by Ilya Pavlov on Unsplash
Vespa Product Updates, May 2021
Advances in features and performance include new int8 and bfloat16 tensor cell types, compact tensor feed format, Approximate Nearest Neighbor using Hamming distance, hash-based attribute dictionaries and case-sensitive attribute search...
Photo by Scott Graham on Unsplash
Vespa Product Updates, March 2021
Advances in Vespa features and performance last month include mass update/delete in /document/v1/, improved memory usage, OR-to-WeakAnd and better full node protection.
Photo by Scott Graham on Unsplash
Vespa Product Updates, January 2021
Advances in Vespa features and performance last month include Automatic Reindexing, Tensor Optimizations and Explainlevel Query Parameter for easier query blueprint tracing.
Photo by Donald Giannatti on Unsplash
Vespa Product Updates, December 2020
Advances in Vespa features and performance include improved tensor ranking performance, Apache ZooKeeper integration, Vespa Python API for researchers and ONNX integration.
Photo by Ilya Pavlov on Unsplash
Vespa Product Updates, October 2020
Improvement to Vespa feeding APIs
Photo by ThisisEngineering RAEng on Unsplash
Vespa Product Updates, September 2020
Introducing ONNX-Runtime, Hamming Distance Metric, Conditional Update Performance Improvements and Compressed Transaction Log with Synced Ack
Vespa Product Updates, August 2020
Introducing NLP with Transformers, Grafana how-to, Improved GEO Search Support, Query Profile Variants Optimizations, & Build on Debian 10
Vespa Product Updates, June 2020
Announcing support for approximate nearest neighbor vector search which can be combined with filters and text search with state-of-the art performance
Vespa Product Updates, May 2020
The May 2020 update includes Improved Slow Node Tolerance, Multi-Threaded Rank Profile Compilation, Reduced Peak Memory at Startup, Feed Performance Improvements, & Increased Tensor Performance.
Vespa Product Updates, April 2020
The April 2020 update includes Top-K hits, smarter data migration and CloudWatch integration. Contributing to Vespa is now easier with the release of a CentOS 7 dev environment.
Vespa Product Updates, February 2020
Advances in Vespa features and performance in February include LightGBM support, improved tensor performance, benchmarking guide and query builder library
Vespa Product Updates, January 2020
The January 2020 update includes information about new tensor functions, updated sizing guides and various performance improvements.
Category internships
Parsing Through the Summer: A Tale of Schemas, Syntax, and Shenanigans
A tale of two interns who dove into the world of Vespa schemas, developing a language server.
A New Bot: Dawn of a Vespa-Powered Slack Renaissance - A Summer Internship Story
Discover how we harnessed the power of Vespa.ai to build an evolving Slackbot that learns and adapts with every interaction!
Summer Internship at Vespa
The tale of 2023's summer interns and their attempt at generating training data for information retrieval with LLMs.
Photo by Arnold Francisca on Unsplash
Summer internship at Vespa
After the summer internship of 2022 the intern has summarized what he has done and his experience at Vespa
Photo by charlesdeluvio on Unsplash
Internship at Vespa
After the summer internship of 2021 the interns have summarised what they have done and their experience at Vespa
Category newsletter
Vespa Newsletter, October 2024
Advances in Vespa features and performance include Global significance models and Nearest Neighbor Search with multiple sparse tensor dimensions.
Vespa Newsletter, September 2024
Advances in Vespa features and performance include Optimized MaxSim with Hamming distance, IDE Support, Pyvespa features, and new notebooks with ColPali examples.
Vespa Newsletter, August 2024
Advances in Vespa features and performance include new pyvespa and Vespa CLI features for queries, feeding, and deployment, shared match-phase heap, Chinese tokenization with Jieba, and new ranking framework features....
Vespa Newsletter, May 2024
Advances in Vespa features and performance include improved vector search performance, fuzzy search with prefix match, RAG, and new Pyvespa and embedding features.
Photo by Scott Graham on Unsplash
Vespa Newsletter, April 2024
Advances in Vespa features and performance include a new SPLADE embedder, float16 support for ONNX models, new Cohere guides, and support for using ColBERT with long texts.
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, February 2024
Advances in Vespa features and performance include the YQL IN operator, new streaming search features, and embed using parameter substitution.
Photo by Scott Graham on Unsplash
Vespa Newsletter, December 2023
Advances in Vespa features and performance include rank score normalization in the global ranking phase, improved feeding using PyVespa, token authentication, and query match debugging!
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, October 2023
Advances in Vespa features and performance include Vespa Cloud Enclave, Lucene Linguistics integration, much faster fuzzy query term matching, and performance and usability improvements.
Photo by Scott Graham on Unsplash
Vespa Newsletter, August 2023
Advances in Vespa features and performance include multilingual models, more control over ANN queries, mapped tensors in queries, and multiple new features in pyvespa and the Vespa CLI.
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, July 2023
Advances in Vespa features and performance include Vector Streaming Search, GPU accelerated embeddings, Huggingface models and a solution to MIPS using a nearest neighbor search.
Photo by Scott Graham on Unsplash
Vespa Newsletter, May 2023
Advances in Vespa features and performance include multi-vector HNSW Indexing, global-phase re-ranking, LangChain support, improved bfloat16 throughput, and new document feed/export features in the Vespa CLI.
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, March 2023
Advances in Vespa features and performance include GPU support, advanced BCP autoscaling, GCP Private Service Connect, and a great update to the e-commerce sample app.
Photo by Scott Graham on Unsplash
Vespa Newsletter, January 2023
Advances in Vespa features and performance include Better Tensor formats, AWS PrivateLink, Autoscaling, Data Plane Access Control and Container and Content Node Performance.
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, November 2022
Vespa features and performance advances include ANN Pre-Filter Performance, Parent Field Hit-Estimates, Model Training Notebooks, and GCP Support.
Photo by Scott Graham on Unsplash
Vespa Newsletter, October 2022
Advances in Vespa features and performance include a BertBase Embedder / model hub, improved query performance, paged attributes, ARM64 support and term bolding in string arrays.
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, September 2022
Advances in Vespa features and performance include rank-phase statistics, detailed rank performance analysis, new query- and trace-applications and a new training video!
Photo by Scott Graham on Unsplash
Vespa Newsletter, June 2022
Advances in Vespa features and performance include ANN with configurable filtering, fuzzy matching, and native embedding support. Also see pyvespa’s new experimental ranking module!
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, April 2022
Advances in Vespa features and performance include tensor and ranking configuration improvements, pyvespa usability features and grouping configuration. Also find new guides for performance and ANN. And a podcast!
Photo by Scott Graham on Unsplash
Vespa Newsletter, January 2022
Advances in Vespa features, performance and operability improvements include: Improved synonym support, faster node recovery, re-balancing and re-indexing, WeakAnd query type and new pyvespa features and sample applications.
Photo by Ilya Pavlov on Unsplash
Vespa Newsletter, December 2021
Advances in Vespa features and performance include tensor performance improvements, match-features and a Vespa IntelliJ plugin.
Photo by Scott Graham on Unsplash
Vespa Newsletter, November 2021
Advances in Vespa features and performance include improved schema inheritance, Intellij plugin for schemas, Hamming distance in ranking, and performance gains in data dump and application deployment.
Category vector search
Photo by Christopher Burns on Unsplash
Query Time Constrained Approximate Nearest Neighbor Search
This blog post describes Vespa's industry leading support for combining approximate nearest neighbor search, or vector search, with query constraints to solve real-world search and recommendation problems at scale.
Category filters
Photo by Christopher Burns on Unsplash
Query Time Constrained Approximate Nearest Neighbor Search
This blog post describes Vespa's industry leading support for combining approximate nearest neighbor search, or vector search, with query constraints to solve real-world search and recommendation problems at scale.
Category Vespa Cloud
Photo by israel palacio on Unsplash