Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

C3.ai's fiscal second quarter results showed promise, but the biggest challenge facing the company is abundantly clear: It has to convert pilots to production to start collecting consumption revenue at scale.

If pilots with customers (prospective or existing) were revenue, C3.ai would be putting up stronger sales gains. Part of the issue with C3.ai's second quarter is expectations. If the company is going to live up to its billing as an enterprise generative AI juggernaut it should have stronger growth.

Another wrinkle is that C3.ai transitioned its revenue model to one based on consumption-based pricing from licensing. That transition equates to more predictable revenue in the future, but there is a transition. C3.ai used to live on large enterprise subscriptions that were lumpy and now has a model more in line with cloud providers and companies like Snowflake.

That transition, which appears to be largely complete, is highlighted in C3.ai's investor deck.

C3.ai reported second quarter revenue of $73.2 million, up 17% from a year ago. Subscription revenue was up 12% from a year ago. C3.ai reported a net loss of 59 cents a share and a non-GAAP net loss of 13 cents a share.

The company projected third quarter revenue of $74 million to $78 million, up 11% to 17% from a year ago.

In many ways, C3.ai is in the right place at the right time. Generative AI has taken off and C3.ai is a key player. I recently detailed a C3.ai project at Baker Hughes illustrating a sustainability generative AI use case (PDF).

Constellation ShortListâ„¢ Artificial Intelligence and Machine Learning Cloud Platforms | C3 AI launches domain-specific generative AI models, targets industries | Get ready for a parade of domain specific LLMs | C3 AI CEO Tom Siebel: Generative AI enterprise search will have wide impact

The challenge is that C3.ai's go-to-market ground game is a work in progress. C3.ai has been building out its partner network with AWS becoming a big channel as the two companies target overlapping industries. C3.ai's qualified pipeline with AWS more than doubled in the second quarter. C3.ai also has partnerships with Google Cloud, Microsoft, Booz Allen and others.

There's clearly interest in what C3.ai can offer. Bookings in the second quarter were up 100% from a year ago, new agreements were up 148% and pilots grew 177%.

Yet, these deals start small and have yet to deliver in full just yet. C3.ai is more a steady climb than a revenue rocket ship at this point. C3.ai launched with massive deals from a small number of customers--often US government agencies and large enterprises--and now is more land and expand due to generative AI.

CEO Tom Siebel said the company is converting pilots to more projects. C3.ai converted two Department of Defense logistics pilots into projects. Siebel said:

"In Q2, we closed 62 agreements, including 36 pilots and trials. Our new pilot count is up 270% from a year ago. Notably, 20 of these were generative AI pilots, a 150% increase from Q1. With the lower entry price points of our pilots, we are more easily able to land new accounts. With our pilots, we are engaging customers across a diverse set of industries in this quarter. Our pilots came from manufacturing, federal, defense, aerospace, pharmaceuticals and other industries."

Over time, C3.ai's business is going to look different. Nearly half of the company's revenue comes from federal, defense and aerospace customers. The pilots and trials underway cover many more industries.

What remains to be seen is how long it takes C3.ai to develop its revenue growth flywheel. Siebel said the sales cycle for generative AI use cases can be as fast as 24 hours with a live application in a month or two with low price points. These customers will ultimately expand their deals as C3.ai delivers value.

"The standard pilot that we have for generative AI and the enterprise is like $250,000. You can get the C3 Generative AI: AWS Marketplace Edition that’s free for 14 days," said Siebel.

Siebel did say that C3.ai is seeing interest in generative AI applications, decisions are taking longer as enterprise put governance around AI. He said:

"Virtually every company in the last 3 to 6 months has created a new AI governance function as part of its decision-making process. These AI governance functions assess and approve those AI applications that will be allowed to be installed in the enterprise. This has candidly added a step to the decision process in AI. You might have heard it here first, but you will be hearing this from every AI vendor in the next few quarters. Take it to the bank."

Overall, Siebel said more AI governance is a good development even if it lengthens the sales cycle. See: The Urgent Case for a Chief AI Officer

Siebel added that the C3 AI Platform will gain traction because it can "solve the disqualifying hobgoblins that are preventing the adoption of generative AI in government, in defense, intelligence, in the private sector."

Those issues include answers from large language models (LLMs) that aren't easily traced and can leak intellectual property and create other liabilities. Enterprises will also move toward an LLM agnostic strategy.

Siebel said:

"I don’t think anybody wants to hook their wagon on to any given LLM today with all the innovation that’s going on in the market and to be dependent on any LLM provider. Our solution is LLM agnostic and addresses every one of those hobgoblins that prevent the installation of generative AI in the enterprise. It took 14 years and $2 billion of software engineering for us to be ready for this."

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

Alphabet's Google has launched Gemini, its most powerful model designed to compete with OpenAI's ChatGPT, in three sizes Gemini Ultra focused on complex tasks, Gemini Pro, an all-purpose model, and Gemini Nano, which is aimed at on-device usage.

The Gemini 1.0 rollout will catch some folks by surprise given that there were reports that Gemini would be pushed to early 2024. With the introduction of Gemini, all three hyperscale cloud providers have announced or upgraded models in recent days. Amazon Web Services outlined Amazon Q at re:Invent. Microsoft is upgrading Copilot with the latest from OpenAI.

"These are the first models of the Gemini era and the first realization of the vision we had when we formed Google DeepMind earlier this year. This new era of models represents one of the biggest science and engineering efforts we’ve undertaken as a company," said Alphabet CEO Sundar Pichai.

In a blog post, Demis Hassabis, CEO of Google DeepMind, said the company set out to make Gemini multimodal from the start. Typically, large language models (LLM) have different modes stitched together.

Hassabis said:

"Until now, the standard approach to creating multimodal models involved training separate components for different modalities and then stitching them together to roughly mimic some of this functionality. These models can sometimes be good at performing certain tasks, like describing images, but struggle with more conceptual and complex reasoning.

We designed Gemini to be natively multimodal, pre-trained from the start on different modalities. Then we fine-tuned it with additional multimodal data to further refine its effectiveness. This helps Gemini seamlessly understand and reason about all kinds of inputs from the ground up, far better than existing multimodal models — and its capabilities are state-of-the-art in nearly every domain."

Google cited a bevy of benchmarks for Gemini and said the model "exceeds current state-of-the-art results on 30 of the 32 widely used academic benchmarks used in LLM research and development."

• AWS, Microsoft Azure, Google Cloud battle about to get chippy
• How much generative AI model choice is too much?

For instance, Gemini Ultra scored 90% on MMLU (massive multitask language understanding), which is based on a combination of 57 subjects, world knowledge and problem solving. Google said its approach to Gemini enables it to think more carefully before answering difficult questions.

In a chart, Google outlined Gemini benchmarks vs. ChatGPT. Gemini 1.0 fared well vs. ChatGPT, but lagged in HellaSwag, a benchmark for commonsense reasoning for everyday tasks. Both models scored about 53% on challenging math problems, which is still better than I'd do. Overall, Gemini 1.0 benchmarks are slightly better than the ChatGPT-4 comparison.

While these benchmarks for models are interesting, enterprises are likely to get a bit of Deja vu with the semiconductor benchmark battles. Read the fine print, conditions and realize that in real-world use cases a slightly better benchmark score may not matter.

According to Google, Gemini Ultra excels in "several coding benchmarks." Coding ability is perhaps the use case with the most returns for LLMs as developer productivity has been a game changer for enterprises.

Google also said that Gemini has built-in safety evaluations including safety classifiers to identify, label and sort out content that's toxic. Gemini Ultra is currently completing trust and safety checks before rolling out broadly. Bard Advanced will launch with Gemini Ultra early in 2024.

Gemini will be used to upgrade Google's Bard and Gemini Nano will power generative AI features on Pixel 8 Pro. 

Infrastructure on Google Cloud

No model is complete without an announcement about training infrastructure and in-house processors.

Google trained Gemini on its in-house Tensor Processing Units v4 and v5e. Google is launching Cloud TPU v5p, which is aimed at training AI models.

Cloud TPU v5p, which powers a pod composed of 8,960 chips and an inter-chip interconnect at 4,800 Gbps/chip.

• Microsoft launches AI chips, Copilot Studio at Ignite 2023
• AWS presses custom silicon edge with Graviton4, Trainium2 and Inferentia2

Those TPUs will be part of Google Cloud's upcoming AI Hypercomputer, a supercomputer that will include integrated hardware, open software and flexible consumption models.

Here's a look at the AI Hypercomputer stack:

Constellation Research analyst Andy Thurai said:

"Google is taking on competitors in three major categories -- OpenAI/Microsoft ChatGPT with their Gemini, AWS and NVIDIA on infrastructure with new TPU chips to become the AI training platform, and IBM/HP/Oracle/ with the AI Hypercomputer.

Gemini was delayed due to concerns about the readiness, safety concerns, and especially the problems with non-English queries. It is clear that Google wants to be careful with Gemini since it will be embedded in multiple products. That caution is another reason the live version got replaced with "virtual demos." But Google can't afford to wait with Gemini and lose mindshare as OpenAI/Microsoft and AWS continually release models.

Gemini has a few differentiators. First, Gemini is multimodal from the ground up. Technically, this LLM could cross the boundary limitations of modalities. Second, Google also released three model sizes rather than one size fits all categories. Third, there are safety guardrails to avoid any toxic content.

Bottom line: Google is trying to become a one stop shop for large enterprises to train their massive LLMs and run on Google Cloud."

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

GitLab is on a $559 million revenue run rate exiting the third quarter with 8,175 base customers. While that quarter is garnering Wall Street attention, enterprises should focus on the competitive dynamics moving GitLab's results as its DevSecOps platform gains traction.

The company reported third-quarter revenue of $149.7 million, up 32% from a year ago, with a net loss of $1.84 a share. Non-GAAP earnings for the third quarter checked in at 9 cents a share.

Gitlab's portfolio includes GitLab Duo, a suite of AI tools, and a platform for agile development, automated software delivery, source code management, security and compliance and continuous integration and delivery.

Constellation Research analyst Holger Mueller said:

"Gitlab is on a path to break half a billion in revenue for the first time in its history, fueled by software powered enterprises. Whether the security angle of its DevOps product is the key driver remains to be seen, as the market and the competition is growing."

Here's a look at why GitLab's third quarter was better than expected.

Jira switching to cloud-only model opens the door. GitLab CEO Sid Sijbrandij said Atlassian Jira customers are evaluating platforms and GitLab is competitive. He said:

"Atlassian’s decision to stop support for its server offering is making customers reconsider what product they use for Enterprise Agile Planning. We are focused on making it easier for these customers to move to GitLab SaaS and self-managed. We recently launched a new Enterprise Agile Planning SKU. Now, GitLab Ultimate customers can easily bring non-technical users into the platform.

Because of that server offering being deprecated, a lot of customers are taking stock of where they're at. It is a natural point for enterprises to evaluate."

AWS and Google Cloud are strong partners and motivated to replace Microsoft's GitHub. Sijbrandij said that it won a third quarter deal with a European telecom against GitHub in partnership with AWS.

With Google Cloud, GitLab is being built into the console. Sijbrandij cited a handful of customer wins against GitHub as enterprises aimed to centralize platforms. Sijbrandij said:

"We've got strong partnerships with both Google and AWS. And they're interested because we help their customers move to their cloud faster. We help them accelerate moving those workloads. And an interesting thing we did with Google recently, we announced at Google Next that they'll be integrating GitLab into their development console for GCP. I think that's a really interesting development that will pay off as many things over the longer term. I think it speaks to the strength of these partnerships together with the AWS example."

• GitHub plans to infuse Copilot throughout its platform

• Constellation ShortListâ„¢ DevOps

• Constellation ShortListâ„¢ AI-Powered Developer Platforms

• Atlassian acquires Loom for $975 million, will add asynchronous video to platform

Compliance and governance. Sijbrandij said security and compliance integration has become a key selling point. GitHub appears to be landing customers like Lockheed Martin and Carfax because it builds security and compliance directly into workflows.

The ability to set controls and governance frameworks that's integrated into developer workflows has become a selling point. Sijbrandij said:

"With point solutions, developers have to wait on security teams to identify vulnerabilities. Or, if they have access to a security scanner to assess their code, they need to manually copy and paste the scanner results back into their development tool. The result is code that isn’t scanned at the time it’s written. This increases the time required to detect and resolve vulnerabilities."

GitLab has GitLab Dedicated, a single-tenant SaaS platform that gives customers data isolation and residency. Sijbrandij added:

“GitLab is the only DevSecOps platform that brings together security, compliance & governance, AI, and enterprise agile planning. Enterprises face complexity from all directions in the form of rapidly increasing user expectations, more advanced cyber-attacks, and more strict industry regulations. We believe they need GitLab to help them navigate this complexity and realize business value. Our platform improves engineering productivity, reduces software spending."

DevSecOps is a critical category as AI workloads grow. Sijbrandij said that AI and integrating it with developer workflows is also driving sales.

"AI needs to be throughout the life cycle and for multiple things, like only 25% of the time of a developer spends on coding, 75% is other tasks. And as developers get more productive, they write more code, you need to also increase the productivity of security and operations. So, we're focused on making it work throughout the life cycle," he said.

Expansion beyond developers. GitLab's enterprise planning suite is attracting usage beyond developers. Business users that have to interact with developers and engineers can use portfolio and project management tools without getting bogged down.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

Microsoft Copilot is being upgraded to OpenAI's GPT-4 Turbo model as the software giant rolls out a series of generative AI updates. The move comes as Microsoft and OpenAI move to put the Sam Altman kerfuffle in the rear-view mirror.

In a blog post, Microsoft said Copilot will soon be able to answer with OpenAI's latest model, which is being tested and will roll out more widely in the weeks ahead.

What remains to be seen is whether these model updates affect usage or become a simple behind-the-scenes rollout. With model choice becoming more of a concern, enterprises are likely to wonder about model updates and what it would take to swap out foundational models if needed.

For now, the Copilot model updates land between the OpenAI soap opera and AWS' launch of its Q generative AI assistant that runs across its services.

• AWS, Microsoft Azure, Google Cloud battle about to get chippy
• How much generative AI model choice is too much?
• Microsoft launches Azure Models as a Service
• Microsoft launches AI chips, Copilot Studio at Ignite 2023

In addition to GPT-4 Turbo upgrades, Copilot will get OpenAI's new DALL-E 3 Model for images and capabilities that combine GPT-4, vision with Bing image search and web search data.

Microsoft also said it is testing Code Interpreter, a service that will provide more accurate calculations, coding, data analysis and visualization. Code Interpreter is in limited testing too. Bing will also get Deep Search, which will use GPT-4 to provide better search results for complex topics.

Also see: How Generative AI Has Supercharged the Future of Work | Generative AI articles | Why you need a Chief AI Officer | Software development becomes generative AI's flagship use case | Enterprises seeing savings, productivity gains from generative AI | Work in a generative AI world will need critical, creative thinking

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

IBM and Meta have formed the AI Alliance with 50 founding members aiming to provide "open and transparent innovation" in artificial intelligence. The catch is that many of the big names--Google, Nvidia and Microsoft--driving the AI advances so far aren't among the founding members.

The concept behind the AI Alliance is notable in that the group wants to bring international players, academics, researchers, governments and developers together to provide an open science and technology ecosystem. The tension around technologies like generative AI revolves around for-profit motives vs. doing what's right for humanity. Security expert Bruce Schneier just penned a long missive about AI and trust worth a read.

And IBM and Meta have a history of contributing to open source as well as initiatives such as Open Compute Project. However, it remains to be seen what the AI Alliance ultimately accomplishes without some key AI leaders.

Also see: How Generative AI Has Supercharged the Future of Work | Generative AI articles | Why you need a Chief AI Officer | Software development becomes generative AI's flagship use case | Enterprises seeing savings, productivity gains from generative AI | Work in a generative AI world will need critical, creative thinking

The founding members and collaborators include the following: AMD, Anyscale, CERN, Cerebras, Cleveland Clinic, Cornell University, Dartmouth College, Dell Technologies, EPFL, ETH, Hugging Face, Imperial College London, Intel, INSAIT, Linux Foundation, MLCommons, MOC Alliance operated by Boston University and Harvard University, NASA, NSF, Oracle, Partnership on AI, Red Hat, Roadzen, ServiceNow, Sony Group, Stability AI, University of California Berkeley, University of Illinois, University of Notre Dame, The University of Tokyo, Yale University and others.

Constellation Research analyst Andy Thurai was quick to note who was missing. He said:

"Major A list players such as AWS, Databricks, Snowflake, Dataiku, Data Robot, Domino Data Lab,  Scale.AI, NVIDIA, Microsoft, OpenAI, Google, Anthropic, Cohere, AI 21 Labs and many others are missing from the list."

Like any alliance, federation or task force by any other name, the jury is out until something actually happens. The AI Alliance indicated that it will be action oriented and focus on "fostering an open community and enabling developers and researchers to accelerate responsible innovation in AI while ensuring scientific rigor, trust, safety, security, diversity and economic competitiveness."

Focus areas for AI Alliance include:

• Developing benchmarks and evaluation standards for responsible development of AI systems as well as tools for safety, security and trust.
• Advance an ecosystem of open foundation models with diversity on many fronts.
• Create an AI hardware accelerator ecosystem.
• Support AI skills and research.
• Develop educational content and resources.

Meta and IBM's experience in open source as well as open hardware could come in handy for kicking off AI Alliance. Simply put, the group is worth watching pending further developments.

Thurai said it's unclear what AI Alliance will ultimately accomplish. He said:

"There are many tools and software components, both open source and commercial already available focused on AI Alliance initiatives. Depending on what is measured there are standards out there. For example, tools like GAIA, MLPerf, and other tools to measure F1 score, model performance, accuracy, and drift are available. I am not sure what the alliance is offering at this point. There are some promises made which are yet to be seen. Tools like Fairlearn, AI fairness 360, Accenture Fairness Tool, etc. to measure model fairness are already available. Some popular benchmarking tools include DAWNBench, GLUE, and SQuAD.”

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

IBM launched its IBM Quantum System Two as well as its next-generation quantum processor as the company touts what's possible today with quantum computing.

At the company's IBM Quantum Summit in New York, Big Blue launched IBM Quantum Heron, a 133-qubit processor that's designed to deliver performance and low error rates relative to its predecessor.

Three Heron processors will power the IBM Quantum System Two, which is the company's first modular quantum computer. Modular quantum processors are being pursued by a bevy of industry players as quantum computing is likely to blend with supercomputing at first. Quantum System Two is located in Yorktown Heights, NY.

IBM also outlined its development roadmap to 2033 with a focus on the quality of gate operations, quantum circuits and running at scale.

Constellation Research CEO Ray Wang said:

“Quantum computers need quantum chips and IBM Quantum System Two addresses the need for faster and more precise error mitigation and correction before we achieve mass adoption. The path to the Blue Jay system by 2033 has been one of IBM's most consistent achievements in computing.”

IBM demonstrated how its 127-qubit IBM Quantum Eagle processor can power systems used as scientific tools to solve chemistry, physics and materials problems.

With IBM Quantum Heron, IBM is planning to use a 5x performance boost with improved error rates to build out a fleet of systems. These systems will mostly be available to customers via cloud platforms. In many respects, quantum computing is entering a more accessible era. 

• AWS launches Braket Direct with dedicated quantum computing instances, access to experts
• Will IonQ make quantum computing enterprise relevant in 2025?
• Quantum Computing Platforms
• Quantum Computing Software Platforms
• Quantum Full Stack Players
• Why your quantum computing vendors are going to look familiar

Meanwhile, IBM outlined a Quantum Development Roadmap with future processor plans and systems. Here's the roadmap (larger format).

IBM also detailed a software stack that will use generative AI and watsonx to make quantum software programming easier. IBM is taking Qiskit 1.0 and adding Qiskit Patterns, which enables developers to create code more easily. Combined with Qiskit Runtime and Quantum Serverless and IBM is looking to create an entire quantum computing stack for multiple scenarios.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.  Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and…...

Read more

MongoDB announced the general availability of MongoDB Atlas Vector Search and MongoDB Atlas Search Nodes in the latest in a series of announcements to enable developers to more easily build generative AI applications.

The general availability of MongoDB Atlas Vector Search and Atlas Search Nodes follows up from a June launch where the company set out to enable AI and large language model (LLM) workloads.

Customers using MongoDB Atlas Vector Search and Atlas Search Nodes include AT&T Cybersecurity, Pathfinder Labs and UKG.

Since that launch MongoDB has had a steady cadence of launches. For instance, at AWS re:Invent, the company said it would integrate MongoDB Atlas Vector Search with Amazon Bedrock and announced it would optimize Amazon CodeWhisperer suggestions for MongoDB developers.

MongoDB steps up generative AI rollout across platform

Those horizontal services are also being complemented by MongoDB's focus on industry specific use cases for healthcare, public sector, manufacturing and automotive as Atlas ramps.

MongoDB, along with companies such as Databricks and Snowflake, are aiming to help developers and enterprises leverage real-time data to build generative AI applications. The argument is that AI applications can't be built efficiently without strong data strategies. Atlas Vector Search and Atlas Search Nodes are designed to give enterprise the capability to search real-time data efficiently.

Key points about MongoDB's two additions:

MongoDB Atlas Vector Search, which is an integrated vector database for MongoDB. Developers can use one API to build generative AI applications across AWS, Microsoft Azure and Google Cloud without duplicating and synchronizing data. MongoDB Atlas Vector Search also enables enterprises to use retrieval-augmented generation (RAG) with pre-trained foundation models.

Atlas Vector Search integrates with models from LangChain, Cohere, OpenAI, LlamaIndex, Hugging Face and Nomic.

MongoDB Atlas Search Nodes provide dedicated infrastructure to manage workloads using MongoDB Atlas Vector Search and Atlas Search independent of operational nodes of the database. This workload isolation performs better at scale and optimizes costs.