MLJAR Studio Changelog

Follow every release in one place. Browse new features, improvements, and fixes for MLJAR Studio.

Latest release

Version 1.2.1

May 20, 2026

MLJAR Studio 1.2.1 adds one-click sharing for AI Data Analyst conversations, one-click deployment for Mercury notebook apps, and several UX improvements across the Get Started and startup flows.

Read notesDownload Studio
  • Added one-click sharing for AI Data Analyst conversations published on the user's personal page
  • Added one-click deployment of Mercury notebook web apps to platform.mljar.com
  • Improved the Get Started view with clearer paths for AI Data Analyst, AutoLab, and AI-assisted notebooks
  • Added startup directory selection and recent files in MLJAR Studio

Release History

Every MLJAR Studio release, sorted from newest to oldest.

Version 1.2.1

Latest

May 20, 2026

Conversation sharing, Mercury app deployment, and smoother startup

MLJAR Studio 1.2.1 adds one-click sharing for AI Data Analyst conversations, one-click deployment for Mercury notebook apps, and several UX improvements across the Get Started and startup flows.

  • Added one-click sharing for AI Data Analyst conversations published on the user's personal page
  • Added one-click deployment of Mercury notebook web apps to platform.mljar.com
  • Improved the Get Started view with clearer paths for AI Data Analyst, AutoLab, and AI-assisted notebooks
  • Added startup directory selection and recent files in MLJAR Studio
Read notes

Version 1.1.4

April 10, 2026

Conversational Notebook reliability, reasoning visibility, and workflow stability

MLJAR Studio 1.1.4 improves Conversational Notebook with visible reasoning support across providers, stronger workflow execution reliability, and multiple auto-fix and UX fixes.

  • Added visible reasoning support for Ollama chat and OpenAI Responses API reasoning summary flow
  • Fixed workflow progression for mixed code/no-code turns and improved startup prompt retention with retry
  • Hardened auto-fix behavior with retry handling, stricter follow-up instruction, and lifecycle logging
  • Fixed Python fenced block parsing for responses containing comment lines
  • Fixed hover accessibility for user-message copy/delete action buttons
  • Added sidebar disclaimer under prompt input: AI can make mistakes. Please verify responses.
Read notes

Version 1.1.2

April 8, 2026

LLM retries and workflows narrow layout fix

MLJAR Studio 1.1.2 adds automatic LLM retries for more reliable responses and fixes workflows behavior in narrow layouts.

  • Added LLM retries
  • Fixed workflows narrow layout
Read notes

Version 1.1.1

April 8, 2026

AutoLab sidebar fix, prompt quality improvements, and EULA update

MLJAR Studio 1.1.1 improves Conversational Notebook and AutoLab quality with better prompt behavior for ML pipelines and chart follow-ups, plus a fix for AutoLab Experiments overview visibility in the sidebar.

  • Fixed AutoLab Experiments overview in the sidebar
  • Improved global prompt behavior to omit ID columns in ML pipelines
  • Improved follow-up prompt behavior to include plotted values in responses for better insights
  • Updated EULA
Read notes

Version 1.1.0

April 7, 2026

Conversational Notebook database connectors, SQL execution, and workflow reliability upgrades

MLJAR Studio 1.1.0 introduces major database and workflow upgrades in Conversational Notebook, including multi-engine connectors, SQL execution with notebook materialization, schema context controls, and broad reliability and UX improvements.

  • Added Database Connectors in the conversational prompt box with connection test/save flow and active-connection chip
  • Added multi-engine connector support (PostgreSQL, ClickHouse, MS SQL Server, Databricks, Snowflake, MySQL, Supabase) with package checks/install flow and clearer connector forms
  • Added SQL execution integration with notebook variables (`df_sql_n`) and direct SQL output rendering in chat
  • Added schema and table context controls, including table scan, table selection for LLM context, and Database Tables visibility in Data Awareness
  • Improved workflow execution reliability and UX, including prompt sequencing, duplicate-send prevention, progress behavior, and Run all/Stop all controls
Read notes

Version 1.0.7

March 27, 2026

Markdown rendering reliability, structured AutoML prompting, and AutoLab scoring consistency

MLJAR Studio 1.0.7 improves conversational notebook rendering and markdown reliability, updates AutoML guidance toward `report_structured()` usage, and refines AutoLab prompting for more stable metric interpretation across iterations.

  • Improved conversational notebook rendering and markdown reliability, including safer handling of currency-style `$` text
  • Updated prompts and guidance so generated MLJAR AutoML examples use `report_structured()` in a notebook-friendly way
  • Refined AutoLab experiment prompting to improve metric interpretation consistency
  • Reduced unstable scoring behavior across AutoLab iterations and added additional chat UX robustness polish
Read notes

Version 1.0.6

March 26, 2026

Chat deduplication stability and conversational notebook execution hardening

MLJAR Studio 1.0.6 is a stabilization release focused on chat reliability in the Electron app, with stronger websocket/message deduplication and safer conversational notebook execution behavior.

  • Improved websocket/message deduplication to prevent duplicated user prompts and streamed assistant responses
  • Improved Electron chat reliability where sidebar messages could previously appear multiple times
  • Tightened conversational notebook execution plumbing to reduce malformed auto-fix loops from parser/sanitization edge cases
  • Reduced chat-flow instability and made day-to-day behavior more predictable and production-safe
Read notes

Version 1.0.5

March 25, 2026

Provider-aware AI access fixes, chat stability hardening, and AutoLab example update

MLJAR Studio 1.0.5 focuses on AI access correctness and runtime stability. This release fixes provider-aware entitlement handling for Ollama/OpenAI paths, removes legacy launcher fallbacks, hardens chat send and frontend initialization behavior, and updates the AutoLab House Prices example dataset URL.

  • Fixed provider-aware AI access gating for non-MLJAR providers to avoid incorrect "AI subscription required" behavior
  • Fixed launcher command routing by removing legacy fallback commands that could trigger wrong entitlement paths
  • Improved chat send stability with duplicate-send protection at input and transport levels
  • Added frontend core singleton guard to prevent duplicate initialization and repeated websocket clients/prompts
  • Updated AutoLab House Prices example dataset URL to `data_train.csv`
  • Improved Studio AI guard messaging and local-provider entitlement logic (`ai_local_provider_access`)
Read notes

Version 1.0.4

March 24, 2026

Structured AutoML reports and NumPy 2 compatibility

MLJAR Studio 1.0.4 upgrades AI Data Scientist reporting with structured, LLM-friendly AutoML output and updates dependency compatibility so mljar-supervised works with modern NumPy versions.

  • Added structured AutoML reports optimized for LLM-friendly output
  • Improved AI Data Scientist report format for more consistent machine-readable sections
  • Removed the `numpy==1.26.4` dependency pin; mljar-supervised now works with NumPy > 2.0
Read notes

Version 1.0.3

March 23, 2026

Stronger runtime stability, AutoLab data setup, and smarter run summaries

MLJAR Studio 1.0.3 focuses on reliability and flow quality across conversational notebooks and AutoLab experiments. This release adds stronger timeout/cleanup guards, improves experiment data and results-directory UX, introduces LLM-based run summaries, and fixes multiple state, session, and startup edge cases.

  • Added Ollama Cloud support for remote model connectivity
  • Added watchdog and timeout guards to prevent hanging "Thinking" and "Checking packages" states
  • Improved AutoLab Step 2 data setup with desktop-friendly multi-file selection and persisted `data_files`
  • Upgraded run summary generation to LLM-based synthesis and standardized report delivery via `run_summary`
  • Fixed notebook rename conflicts and prevented cross-notebook variable leakage in conversational workflows
Read notes

Version 1.0.2

March 16, 2026

Stabilized conversational UX, AutoLab experiment flow, and release/version workflow

MLJAR Studio 1.0.2 focuses on stability and polish across conversational notebooks, AutoLab experiment UX, and version/release flow, with multiple reliability and usability fixes in sidebar and run-state interactions.

  • Switched conversational prompt context to interleaved notebook format with paired code and outputs
  • Improved AutoLab run-state and stop-experiment UX with clearer controls and safer actions
  • Fixed sidebar chat regressions including Clear chat and Fix with AI response handling
  • Polished thinking UI, scrolling behavior, and large-variable column summarization
Read notes

Version 1.0.1

March 14, 2026

Better conversational notebook tooling and AutoLab polish

MLJAR Studio 1.0.1 improves Variables Inspector and Package Manager styling, adds support for conversational notebooks in both tools, fixes open code issues in conversational notebooks, and brings UI and assistant fixes across AutoLab and the classic notebook experience.

  • Improved styles for Variables Inspector and Package Manager
  • Added conversational notebook support in Variables Inspector and Package Manager
  • Fixed open code issues in conversational notebooks
  • Better AutoLab Experiments UI and fixes for AI Assistant in classic notebooks
Read notes

Version 1.0.0

March 5, 2026

Conversational notebooks, AutoLab experiments, and flexible AI providers

MLJAR Studio 1.0.0 introduces conversational notebooks with your own AI Data Analyst, AutoLab experiments for autonomous machine learning optimization, and flexible LLM provider settings with support for MLJAR, OpenAI, and Ollama-based open models.

  • Added conversational notebooks with AI Data Analyst
  • Added AutoLab experiments for autonomous ML pipeline optimization
  • Added external LLM provider configuration
  • Support for MLJAR, OpenAI, and Ollama-based models
Read notes