Stories by DISCURSIVE

Cutting LLM Batch Inference Time by Half with Dynamic Prefix Bucketing

Agentic systems are just query engines for unstructured data

Benchmark: Spark vs. Ray Data vs. Daft on Multimodal Workloads

Show HN: Flotilla, the Distributed Engine for Multimodal Pipelines

Profiling multimodal workloads: lessons from Daft

Processing 24T tokens for LLM training with 0 crashes (what made it possible)

We Hit 100% GPU Utilization–and Then Made It 3× Faster by Not Using It

Elasticsearch Was Great, but Vector Databases Are the Future

Have you faced any challenges of using Elasticsearch for vector search?

BGE-M3 vs. Splade

Enhance information retrieval with learned sparse embeddings

The emerging trend of vector databases

The Future of Vector Database

Milvus Supports NVDIA's Graph Index

Zilliz Cloud Jan 2024 Launch

OpenAI RAG vs. Your Customized RAG: Which One Is Better?

Demystify Benchmark Result Divergence: Milvus vs. Qdrant

Pipelines for Unstructured Text Data

DiskANN: A High-Performance Index for Managing Large Vector Volumes

Kubernetes Tutorial: Monitoring HTTP Service Health

1