RAG
Retrieval-Augmented Generation (RAG) is a methodology that optimizes the output of a Large Language Model (LLM) by referencing an authoritative, external knowledge base or Vector Database before generating a response. RAG helps models access real-time information and drastically reduces hallucination.
Frequently Asked Questions
How is RAG different from fine-tuning?▼
Fine-tuning modifies the internal weights of the model, which is expensive and slow. RAG acts like an open-book exam, passing relevant documents directly into the prompt context window.
Does RAG require retraining a model?▼
No, RAG works with pre-trained models by retrieving documents and appending them to the prompt.
Quick Facts
- CategoryInformation Retrieval
- Key ApplicationEnterprise search, dynamic question answering, and customer support
Coverage Trend12 Weeks
Related AI Terms
RAG Media Coverage & Intelligence
Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why
Patient contexts span hundreds of heterogeneous documents and thousands of structured data points, yet the document-level metadata that AI systems need for retr
Toten: Knowledge-Based Ontological Tokenization Of Physical Quantities And Technical Notation In Brazilian Portuguese
Byte-Pair Encoding tokenization is statistically efficient for vocabulary compression, but semantically blind to structured technical entities, fragmenting phys
Measuring Curriculum Alignment across Topical Coverage, Competency, and Cognitive Depth: A Longitudinal Framework Applied to CS2013 and CS2023
Undergraduate computer science is governed by international curricular guidelines revised about once a decade, yet programs lack a reliable, reproducible way to
Everpure launches Enterprise Data Cloud blueprint to guide AI data strategy
Fragmented data, siloed infrastructure and reactive portfolios are problems that every enterprise has to deal with, and fixing them efficiently requires a completely new operating model - one that the Enterprise Data Cloud is designed to deliver. That's the basis for Everpure's new Success Blueprint
Everpure and WWT say data-ready AI infrastructure starts with clean, governed data
Getting to production-ready AI requires much more than fast storage because enterprises need data-ready AI infrastructure that's built on clean, governed and well-understood data before any meaningful deployment can scale. It's a shift that's redefining what partners bring to the table, according to
Three insights you might have missed from theCUBE's coverage of Broadcom's 'Modern Private Cloud' event
Enterprises are rapidly pushing AI projects into production in the race for higher revenues - but it's not without cost and security risks. Private cloud is emerging as a bedrock for AI infrastructure as businesses seek control, cost predictability, security, and compliance. Broadcom Inc. is positio
Three insights you may have missed from theCUBE's coverage of FinOps X
AI costs are becoming one of the most difficult aspects of enterprise AI adoption. Unlike traditional cloud or software-as-a-service spend, AI costs are shaped by dynamic usage patterns, model behavior and external interactions, making it harder to keep investments aligned with business value. As en
New CoreWeave SUNK Capabilities Help Teams Build Modern AI Research Clusters
Why SUNK matters for teams building modern AI research clusters: a faster path to productive training, less compute fragmentation, and deeper operational visibility.
NEA's Tiffany Luck says enterprises are still figuring out their AI ROI
Tokenmaxxing was the hottest trend in Silicon Valley earlier this year, with CEOs encouraging employees to push AI usage as far as it would go. Then the bill ca
NEA's Tiffany Luck on AI IPOs, personal agents, and the ROI reckoning
Tokenmaxxing was the hottest trend in Silicon Valley earlier this year, with CEOs encouraging employees to push AI usage as far as it would go. Then the bill ca
Everpure accelerates AI workloads with Data Stream and unveils data-primacy architectural vision
Big-data storage company Everpure Inc., formerly known as Pure Storage, is rethinking enterprise data architectures to facilitate better access and scalability for artificial intelligence workloads. At its annual customer conference, Pure Accelerate, the company today announced the immediate availab
Nexla's Express solution leverages conversational interface to fuel agentic AI
Working with complex enterprise data once required extensive technical expertise. That barrier is lowering, as demonstrated by data integration platform provider Nexla Inc.'s launch of a conversational data engineering platform designed to make the process more accessible. In November 2025, Nexla un
Qualcomm takes spatial computing into the AI era with Snapdragon Reality Elite
Qualcomm Technologies Inc. today unveiled a brand-new chip featuring the Snapdragon Reality Elite, a rebrand of its existing line of powerful silicon designed to power next-generation immersive virtual reality and mixed reality experiences. Reality Elite is the successor to the XR2+ Gen 2, announced
Xreal unveils Aura, its lightweight smart glasses powered by Android XR
Lightweight augmented reality glasses maker Xreal Inc. officially unveiled its Aura smart glasses today, powered by Android XR and built with Qualcomm Inc.'s Snapdragon. Formerly known as Project Aura, these new smart glasses will launch this Fall, combining lightweight, see-through wired extended r
Glean's AI platform leverages enterprise data to power models and agents
Glean Technologies Inc.'s roots are in enterprise search, which has provided the company with a useful springboard to become one of the leading developers of enterprise-grade autonomous agents and AI-powered business solutions. The company's ability to help businesses create "horizontal" AI agents t
Mobileye is entering the US robotaxi market with standalone service
The service will leverage its Moovit platform to launch in an a US city in 2027.
PixelRAG beats text parsers on accuracy and cuts AI agent token costs 10x
Most enterprise RAG pipelines start the same way: a text parser converts web pages and documents into plain text so they can be chunked and indexed for retrieva
Three insights you may have missed from theCUBE's coverage of Snowflake Summit 2026
If the first wave of enterprise artificial intelligence was about compute and foundation models, the next is shaping up to be about the software and data infrastructure needed to make those models useful in real businesses. The first AI winners sold compute: graphics processing units, servers, netwo
How memory tools can make AI models worse
New research suggests that AI memory systems can degrade model performance and encourage sycophantic tendencies.
GM Energy introduces V2G support and new energy storage battery chemistry
There are more than a quarter of a million V2G-capable GM EVs on the roads already.
Unlocking dependable responses with Gemini Enterprise Agent Platform's Agentic RAG
Data Management
Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation
Multi-step agentic retrieval-augmented generation (RAG) pipelines have demonstrated significant capability for c
Uber caps employee AI spending after blowing through budget in 4 months
Uber's cutback has occurred after the company had reportedly encouraged staff to use AI as much as possible.