MLJAR Studio
mljar.com
- Free plan
- Windows
- macOS
- Linux
- Self-hosted
What MLJAR Studio is
MLJAR Studio is a Python workspace for data analysis, machine-learning experiments, and notebooks, designed for local, offline-first work. Ask questions in plain language and it generates Python analysis that remains visible and editable. Its experiment agent can tune models, identify useful features, compare models, track experiments, and create reports. It works with CSV, Excel, and Parquet files, as well as SQL databases including PostgreSQL, ClickHouse, and MySQL. Notebooks can be turned into interactive Mercury web apps hosted on your own infrastructure. Data processing is local by default; an external provider or MLJAR AI cloud add-on can change that. With OpenAI configured, prompts and relevant context may be sent to that provider under your settings and its terms. Studio runs on Linux, macOS, and Windows, and can be self-hosted. The Free plan includes 50 prompts per month and one public Mercury app; paid options include Pro at $20 a month, Business at $60 a month, and a $199, paid once perpetual license with one year of updates.
Who it suits
MLJAR Studio suits analysts, data scientists, and teams working with data, from beginners to experts. It is also suited to people who want to turn notebooks into web apps hosted on their own infrastructure.
Good to know, and what to watch
- Generates editable Python analysis from plain-language questions
- Supports CSV, Excel, Parquet, and SQL databases
- Tracks and compares machine-learning experiments
- Notebook apps can run on your own infrastructure
- Data processing is local by default
- Free tier allows only one public Mercury app
- Free tier has no private Mercury web apps
- Perpetual license includes one year of updates
- External AI providers may receive prompts and context
Plans
Every plan and priceFree
Pro
Business
Perpetual license
Free plan
Getting started
- Open https://mljar.com/studio/
- Choose the Free, Pro, Business, or perpetual-license option
- Install Studio on Linux, macOS, Windows, or a self-hosted environment
- Configure an AI provider such as MLJAR AI, OpenAI, or Ollama
- Import CSV, Excel, Parquet, or supported SQL data
- Create a notebook or ask for analysis in plain language
Limits to check
Free includes 50 prompts per month, 10 published conversations, and one public Mercury web app. It has no private Mercury web apps; Pro and Business raise quotas and add private apps.
The verdict
MLJAR Studio brings analysis, machine-learning experiments, and Python notebooks into one workspace, with local processing as the default. Check the plan limits and provider settings if you need private published apps or use an external AI service.
Questions people ask
Is MLJAR Studio free?
Yes. The Free plan costs nothing and includes core Studio features. Paid plans start at $20 a month.
Which operating systems does it support?
Studio runs on Linux, macOS, Windows, and self-hosted environments.
Which data sources can it use?
It works with CSV, Excel, and Parquet files, plus PostgreSQL, ClickHouse, MySQL, and other SQL databases.
Can processing stay local?
Data processing is local by default. External transmission occurs when an external provider or MLJAR AI cloud add-on is configured.
Can I publish notebook apps?
Yes. Studio turns notebooks into Mercury web apps hosted on your infrastructure.
What the maker publishes
- AI data analysis
- Users can ask questions in plain language and Studio generates and runs Python analysis while keeping the code visible and editable. source
- AI providers
- Documented provider options include built-in MLJAR AI, OpenAI, Ollama local, and Ollama Cloud; the product page also names Anthropic and OpenAI-compatible providers. source
- Audience
- The maker describes Studio as suitable for beginners and experts and says it serves analysts, data scientists, and teams working with data. source
- AutoML
- Its experiment agent can tune models, discover useful features, compare models, track experiments, and generate reports. source
- Cloud provider caveat
- When OpenAI is configured, prompts and relevant context may be sent to OpenAI according to the user's configuration and provider terms. source
- Company
- MLJAR identifies its founders as Aleksandra Płońska and Piotr Płoński on its About page. source
- Data sources
- The AI Data Analyst page says Studio works with CSV, Excel, and Parquet files, plus SQL databases including PostgreSQL, ClickHouse, and MySQL. source
- License updates
- The perpetual license includes one year of updates, and newer releases after that period require an update renewal. source
- Notebook apps
- Studio can turn notebooks into interactive web apps powered by Mercury and host them on the user's own infrastructure. source
- Platforms
- The download page offers installers for Windows, macOS, and Linux. source
- Privacy
- Studio is designed for local, offline-first use; its AI policy says data processing is local by default and no data is transmitted externally unless an external provider or MLJAR AI cloud add-on is configured. source
- Publishing limits
- The Free tier includes one public Mercury web app and no private Mercury web apps; Pro includes one private app and Business includes three. source
- Purpose
- MLJAR Studio is a local AI data workspace for data analysis, machine learning experiments, and Python notebooks. source
Checked against the maker's pages on 3 Oct 2026.