Most Downloaded AI Model Isn't What You Think, according to Liftr Insights data

Tabular data, not NLP, has more downloads per model on average, as shown by
Liftr Insights data

AUSTIN, Texas, March 5, 2025 /PRNewswire/ -- Liftr Insights, a pioneer in market intelligence driven by unique data, released an article about Tabular AI Models as part of its AI Series showing that Tabular Models are downloaded 8 times more than natural language models ("NLP"). This is despite the fact that NLP is what people think of when they think of AI models, thanks to the press regarding OpenAI, DeepSeek, Gemini, and others.

Tabular models are designed to work well with structured data sets of any kind. They can be used for forecasting or filling in missing data.

In addition to providing objective data about Tabular models, the article also provides supporting data about Multimodal, Computer Vision, and Audio, as some of the more popular types, in addition to the most popular and commonly known, NLP.

"As we have been exploring a wealth of AI information for some clients, the data has uncovered many interesting signals," says Tab Schadt, CEO of Liftr Insights. "This is in addition to interesting trends we're seeing."

As stated in the Tabular article, "Tabular models have only 1.0% of the total NLP downloads, but the average number of downloads per model is over 24,000." The bulk of these downloads, as shown by the data, are due to Chronos models.

The article about Tabular models also delves into the sizes of models, since there are often misconceptions that, well, size matters.

"This is powerful data about AI managing data," says Schadt. "This is the type of information that helps Market Intelligence analysts who want objective, reliable data to get beyond the hype of AI."

The article can be found at https://liftrinsights.com/series/ai-model/ai-distills-millions-of-data-records and is part of a series of articles about AI.

About Liftr Insights
Liftr Insights generates reliable market intelligence using unique data. Liftr Insights cloud and semiconductor data cover the globe and represent census data for over 75% of the public cloud. Liftr Insights subject matter experts translate company-specific service provider data into actionable alternative data, including details about configurations, components, deployment geo, and pricing for:

• Server processors: Intel Xeon, AMD EPYC, Aliyun Yitian, AWS Graviton, Azure Cobalt, Google Axion, NVIDIA Grace, and Ampere Computing Altra
• Datacenter compute accelerators: GPUs, FPGAs, TPUs, and AI chips from NVIDIA, Intel, AMD, AWS, Google, and Qualcomm
• Public cloud AWS, Microsoft Azure, Alibaba Cloud, Google Cloud, Oracle Cloud, Tencent Cloud, CoreWeave, Lambda, and Vultr.

For a sample of Liftr data, check out the Liftr Cloud Regions Map at https://bit.ly/LiftrCloudRegionsMap. 

Liftr and the Liftr logo are registered service marks of Liftr Insights. The following are trademarks and/or service marks of Liftr Insights: Liftr Insights, Cloud Components Tracker, Intelligence Compute Tracker, and Liftr Cloud Regions Map. 

The following are registered intellectual property marks, trademarks, or service marks of their respective companies:
Amazon Web Services
Microsoft Azure
Alibaba Cloud
Google Cloud
Oracle Cloud
Tencent Cloud
CoreWeave
Lambda
Vultr
Intel Corporation
Ampere Computing
Qualcomm
NVIDIA
AMD
ARM
Chronos

SOURCE Liftr Insights