Dataiku Review, Pricing & Alternatives (2026)

Dataiku is an enterprise data science and AI platform founded in 2013 by Florian Douetteau, Clément Stenac, Thomas Cabrol and Marc Batty. The company says more than 500 companies use it. Its December 2022 Series F raised $200 million at a $3.7 billion valuation, bringing primary funding to roughly $600 million. Dataiku offers hosted and self-hosted paid editions, plus an installable Free Edition for smaller projects. Reuters reported in October 2025 that the company hired Morgan Stanley and Citigroup to prepare a possible United States IPO. That report described preparation for a potential transaction, not a completed listing. Dataiku’s defining product choice is a shared workspace for analysts, engineers and data scientists, combining visual work with programming and controls for production delivery.

A typical Dataiku project begins with data connections and preparation, using visual recipes or code to clean, join and transform information. Users can then work in notebooks, Python, R, visual machine learning and AutoML, before publishing dashboards or applications. Scenarios schedule repeatable jobs, while model deployment and operations features support production use in paid editions. Governance controls help organizations manage projects and models across teams. LLM Mesh connects projects to external model providers through customer or provider credentials. The interface accommodates less technical users, while Python and R give experienced practitioners direct coding options. This combination lets one project move from exploratory preparation to modeling, reporting and operational workflows without requiring every participant to use the same interface.

Dataiku DSS supports Java 17 or 21, Python 3.11, 3.12 or 3.13, and R 4.x according to current installation documentation. Its LLM connection documentation lists OpenAI models including GPT-6.1 Sol, GPT-6 Astra, GPT-6 Luna, GPT-5.6 Terra, GPT-4.1, o1, o3 and o4-mini, along with provider connections through services such as AWS Bedrock. These connections use customer or provider credentials rather than a model bundled by Dataiku. DSS 14.2 release notes dated October 17, 2025 list Model Context Protocol support, AI Portfolio automated tagging, and additions for classical machine learning and time-series forecasting. The release information shows Dataiku extending both conventional analytics and newer model integration, while keeping deployment, governance and workflow management inside DSS.

Dataiku does not publish dollar prices for paid editions. Those editions are quote-based, hosted by Dataiku or self-hosted, and add enterprise collaboration, governance, operations, model deployment, automation, security, LLM Mesh and larger-scale usage. The installable Free Edition works on Mac, Linux or a virtual machine and supports up to three users, data preparation and basic data projects or applications. It excludes deployment, automation and governance and has limited connectivity. The managed Cloud trial lasts 14 days and is normally provisioned in under two minutes. Dataiku publishes limits of two users in the trial space, one API service, four CPUs and 32 GiB of elastic compute. The trial excludes Govern and advanced LLM Mesh, then deactivates.

The main limitation is the gap between Dataiku’s accessible Free Edition and the commercial controls needed for production work. Free users cannot deploy or automate projects, and they do not receive governance features. The Cloud trial offers a broader managed experience, but its 14-day duration and compute limits restrict sustained evaluation. Compared with H2O AI Cloud, Dataiku provides wider collaborative workflow and governance, while H2O-3 is the more permissive open-source engine. Compared with DataRobot, Dataiku exposes more coding and data-pipeline control, while DataRobot can be simpler for packaged enterprise AutoML. Databricks, SAS Viya, Alteryx and KNIME are additional alternatives. Organizations needing governed, collaborative AI delivery across technical and nontechnical users should pick Dataiku.