Vectara Reports Record First-Half Momentum with Surging Revenue and Marquee Enterprise Wins. Over 100% New Revenue Growth in 2026.
Broadcom (NASDAQ:AVGO)
PALO ALTO, CALIFORNIA, CA, UNITED STATES, August 4, 2026 /EINPresswire.com/ — Over 100% New Revenue Growth in 2026 Driven by Incremental Enterprise AI Deployments, Strategic Platform Certifications, and a Stronger Ingestion Engine
Vectara, the trusted enterprise platform for building Agentic and RAG solutions that run on-premise, in VPC, and in SaaS today announced record momentum for the first half of 2026. The company more than doubled its revenue year over year, fueled by a wave of new enterprise deployments, expanded usage at marquee accounts, and a series of strategic AI partner certifications. Enterprises moved agentic AI from experimentation into production and standardized on Vectara to keep those systems accurate, while enabling new applications around failure analysis remediation, contract lifecycle management, and faster manufacturing design and build efficiency.
Growth was broad-based across new customer acquisition and expansion within the existing base, with especially strong demand from semiconductor and technology customers, drawn to Vectara’s hardened, secure platform for handling complex image, table, and text datasets.
In the first half, Vectara welcomed new customers including: a global flash storage manufacturer running an on-premise failure analysis agent for its FA engineers. Our strong multimodal support was key to reducing their FA turn around time.
Another new customer is a large fabless semiconductor supplier with a broad catalog of configurable parts which Vectara is delivering four AI agents: search, chat, part selection, and outbound on one grounded knowledge base, reaching 95% accuracy. Leveraging these four AI Agents, Vectara was selected as the front-end e-commerce interface powering the company’s sales and marketing experience demonstrating how the platform now reaches beyond internal knowledge assistants to drive customer-facing, revenue-generating workflows.
Vectara’s partner ecosystem also produced standout results during the half. The company became a certified AI partner for VMware and earned certification as a certified Agentic solution for VMware Cloud Foundation (VCF) 9.x, positioning Vectara to deliver trusted agentic AI within the VMware environments that the world’s largest enterprises rely on. Separately, Vectara’s Zoom partnership was certified after being built directly from a customer need, turning a real-world requirement into a validated, repeatable joint solution.
“Getting agents to operate and scale correctly in production involves combining multiple complex components in the right way. Vectara has proven the platform can do this well especially when multimodal data is involved. This drives measurable outcomes in Broadcom’s customer experience and silicon engineering.” Alan Davidson, Chief Information Officer, Broadcom
“This was a defining first half for Vectara. Enterprises are no longer asking whether to deploy agentic AI, they are asking what applications can we service that we can bring Vectara’s trust, accuracy, and control to” said Tallat Shafaat, CEO & Founder, Vectara. “New customers like Sandisk and Altera, expansion at one of the largest semiconductor & software companies in the world, and certifications with VMware and Zoom show that our platform is becoming the standard for production-grade, grounded AI. Combined with the investments we’ve made in ease of use and integration flexibility we head into the second half with strong momentum and an even stronger product.”
Platform Advancements: Build, Ingest, Govern, Operate
Underpinning the commercial momentum, Vectara shipped platform advancements in the first half that span the full agent lifecycle: agent-driven data ingestion, plain-language agent building, governed tool integration, and production operations. Every capability arrived API-first, a documented endpoint from day one. Highlights from recent Vectara release notes include:
Agentic ingestion. Vectara Pipelines put a real agent in front of every document: each record gets a fresh agent session that decides how to extract, route, index, or discard it with dead-letter retry and run cancellation managed for you. Pipelines pull from SharePoint, Box, Confluence, Google Drive, S3, Wolken, and websites (sitemap and crawl discovery, JavaScript rendering), and carry the source system’s own permissions through to the index instead of flattening access control. Spreadsheet ingestion and image embeddings bring visual and tabular content into retrieval, and a new bulk metadata API re-tags an entire corpus in one call, no reindexing.
io, the AI assistant for building on Vectara. Describe what you want, e.g. “Make me a scheduled pipeline that ingests my docs into a corpus”, and io inspects your account, drafts the configuration, creates agents, pipelines, and API keys while enabling you to review at every step. Vectara is the most configurable agent platform on the market; io makes every knob reachable in a sentence.
Governed tools. Anyone can bolt tools onto an agent; Vectara governs the tool boundary. Encrypted agent secrets inject credentials only at execution time and never return in plaintext. Every agent runs under its own identity with independently scoped permissions. Input and output transforms reshape what enters and leaves every tool without custom code, reusable tool configurations keep fleets of agents in sync from one definition, and sandboxed lambda tools compose other tools while staying network-isolated and blind to the underlying secrets. On top of that governance layer: new tool suites for GitHub, Jira, and Kubernetes, richer Model Context Protocol support with file, image, and audio results, and agent connectors for Zoom Contact Center and Google Chat alongside Slack.
Production-grade operations. The gap between an agent demo and an agent product is what happens after launch, and it’s where most platforms stop. Agent aliases give callers a stable name with server-side routing policies, enabling canary rollouts and tenant-based routing with zero client redeploys. Agent traces reconstruct every run as a tree of spans (each model call, tool execution, and guardrail check, with timing and token spend) and the new Observability console charts latency percentiles, errors, and tool behavior across your fleet. Context compaction and automatic tool-output offloading keep long-running sessions sharp by default.
About Vectara
Vectara is the trusted enterprise platform for building agentic AI assistants that run on-premise, in your VPC, and in SaaS. Purpose-built to reduce hallucinations and deliver accurate, explainable, and secure results, Vectara combines best-in-class retrieval, generation, and guardrails so that enterprises can confidently move applications into production. For more information, visit www.vectara.com.
Media Contact
press@vectara.com
Press
Vectara
+1 815-751-4899
Press@vectara.com
Visit us on social media:
LinkedIn
YouTube
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
Media gallery
