Webinar

Introducing Mockingbird: A High-Performance Generative Model for RAG

Mockingbird LLM is here, and it is bringing unrivaled accuracy to RAG workloads. As part of the Vectara Trusted GenAI Platform, Mockingbird provides users with generative responses and structured outputs with levels of performance and accuracy that are ideal for running RAG workloads. Vectara platform users can now use the full solution without dependencies on third-party generative services. This makes Mockingbird not only the perfect choice for RAG performance but also versatile enough to be deployed across a multitude of form factors.  

Join Sean Anderson (Head of Product Marketing) and Suleman Kazi (Software Engineer) as they explore several benchmarks that show the true power of the new model. You will get a view of the rigor put into making a generative LLM that enhances the performance of the entire RAG process with enhanced support for structured outputs and hallucination mitigation. Mark your calendars for the live event on August 15th, 2024.

Webinar

Build Enterprise-Grade Generative AI Apps with Your Apache Nifi Data

Thousands of the largest and most secure enterprises use Apache NiFi for real-time data ingestion and cyber-security use cases. Did you know NiFi was originally built at the NSA to handle primarily unstructured data?

NiFi, at its foundation, is built for the data types needed by today’s GenAI Systems. The creators of NiFi have formed Datavolo which allows you to leverage your NiFi skills to rapidly and securely ingest all of your multimodal data into the world’s fastest time-to-market platform for GenAI Apps, Vectara. Come learn how you can leverage your Nifi data to build enterprise-grade AI applications in a few hours with the power of the Datavolo/Vectara. Join Sam Lachterman (Field CTO, Datavolo) and Justin Hayes (Field Engineer, Vectara) on June 18th, 2024.

Webinar

Introducing the Factual Consistency Score: Your AI Evaluator

Recently, Vectra introduced an evolved version of its popular open-source Hughes Evaluation Model (HHEM), which detects the level of hallucinations in popular LLMs and in generated responses from those systems. It’s a calibrated score that helps developers evaluate hallucinations automatically. The new featurre can be used by our customers to measure and improve response quality. The FCS can also be used as a visual cue of response quality shown to the end-users of RAG applications

The Factual Consistency Score is Vectara’s integration of the hallucination detection model into its generative AI platform. Machine Learning Team Lead Forrest Bao will show us the new FCS score but in the Vectara UI and through the API.

In-person & Streaming Event

Deep Dive Into RAG Architectures

📣 Join the Vectara ML team and fellow industry peers for an evening of thought leadership talks 💡 and social networking 🥂 as we do a deep dive into the hottest topic in LLMs – Retrieval Augmented Generation (RAG) Architectures 🔥.

Webinar

Dive into Vectara Chat

Introducing Vectara Chat, a powerful new set of tools for developing an LLM-powered chatbot grounded in your own data. Vectara Chat introduces a revolution in chatbot technology, blending RAG, LLMs, and user-centric design to transform how businesses and individuals communicate. On top of the core RAG technology, Vectara chat offers chat history that incorporates previous chats into context for a more intelligent response. Head of Product Shane Connelly will take us through this exciting new product feature aimed at helping more chatbots reach production.

Webinar

GenAI Revolution: Master Data Ingestion with Airbyte & Vectara

 

 Overview of GenAI Data Ingestion Challenges

 Introducing Vectara and Airbyte Integration

  Setting up Airbyte and Vectara

  Interactive Q&A sessions

 

Webinar

First Look: New Features in Vectara 2024

Get a first glimpse of new enhancements of the Vectara platform aimed at improving Hybrid Search, decreasing hallucinations, and driving toward better relevance and making GenAI applications even easier to build. Head of Product Shane Connelly will take us through several exciting product updates to give you the full picture of the advancement of our GenAI platform.

Webinar

Evaluating Hallucinations in RAG

Hallucinations continue to hinder the adoption of AI systems into production. Companies still do not fully trust that the conversational AI systems they build will not produce inaccurate results and degrade user confidence. Many vendors make claims to all but eliminate hallucinations, but to date, there are not many impartial mechanisms to quantity the amount of hallucination.

We will explore open-source tools that help provide visibility into the level and impact of hallucinations for many models.

Session Replay

SonoSim + Vectara Success Story

On Oct. 16, 2023, SonoSim CEO & Co-Founder, Eric Savitsky, MD, delivered a talk titled “GenAI & Education Technology” at Vectara’s “Boomerang & Ecosystem Launch Event” at Vectara HQ in Palo Alto, CA. The medical community is hypersensitive to current hallucination risk and the context of this 15-minute presentation accurately reflected this balance.

During this informative talk, Dr. Savitsky detailed:

• The importance of multi-modalities in consideration of AI implementation in healthcare
• Why details matter in descriptive text inputs
• Why SonoSim chose Vectara over a DIY approach