Hugging Face tests a 4× reduction with pooling-aware training
Pooling-aware training reduced document vectors to one quarter in a multi-vector search index while retaining 97.8% of the unchanged nDCG@10 performance.
1 min readRead story
CHOOSE YOUR FOCUS
What do you create?
NEWS FEED
RAG
TopicRAGClear topic
Sorted by AI Radar publication time · Times shown in Istanbul (UTC+3).Clear filters
Amazon Bedrock adds single-shot RAG and retrieval-only modes
Amazon Bedrock Managed Knowledge Base now offers one-search answer generation and passage-only retrieval without a model call through the same AgenticRetrieveStream API.
1 min readRead story
Cohere Embed 5 and Rerank 5 arrive on OpenRouter’s unified API
OpenRouter users can call Cohere’s Embed 5 and Rerank 5 models with their existing API key; four new endpoints cover text, image and mixed-content RAG workflows.
1 min readRead story
NVIDIA Nemotron 3.5 gives agents a multimodal document-retrieval pipeline
NVIDIA released open-weight Embed models for finding evidence across text and images, a Rerank model for reordering results, and a Parse model for structuring documents.
2 min readRead story
Redpine gives AI agents licensed access to 20 million research papers
BMJ Group, Sage and four other publishers are bringing more than 20 million peer-reviewed papers into a licensed access layer that AI agents can query through MCP and API.
2 min readRead story
BEYOND THE HEADLINE
What changed?
What can it bring to your work?
Every story includes a concise summary, its creative workflow implications and original sources. Explore official demos and examples alongside the news.
Our editorial approach
