Skip to content

Databricks releases, every cloud

40 releases of 1321

Last updated

  1. Base environments are now supported on classic compute (Beta)

    Manages Python dependencies using Databricks-provided or custom environments.

  2. Genie Code adds conversation controls

    Lets you fork conversations, use /btw for side questions, and edit earlier messages.

  3. Git Folder Serverless is in Beta

    Shares serverless compute and pyproject.toml environment across notebooks and files.

  4. Genie Code scheduled tasks are now generally available

    Runs prompts on a recurring schedule, producing a continuable Genie chat with results.

  5. Genie Code scheduled tasks will soon be available by default for workspaces with the compliance security profile enabled Coming soon

    Replaces manual runs of recurring analysis tasks. Available in October 2026.

  6. Genie Code for AI Runtime (Public Preview)

    Generates distributed training code and resolves environment issues. Requires AI Runtime.

    AWS, read it on their docsAzure, read it on their docsGCP(not on this cloud)SAP(unknown)
  7. Run the change, do not just read it. Hands-on labs in your own Databricks workspace, graded when you submit. Browse labs
  8. Genie Code is now available as a Lakeflow Jobs task (Beta)

    Runs prompts autonomously, reading upstream outputs and calling tools as needed, returning a conversation link.

  9. Default Python package repositories are GA for Lakeflow pipelines and classic compute

    Supports private or authenticated registries as workspace defaults. Applies to UI and API created classic compute.

  10. Choose an effort level in Genie Code

    Sets balance between response quality and cost, defaults to Auto for highest quality.

  11. Manage project environments with %uv commands

    Requires environment version 5 or above. Supports serverless notebooks.

  12. Genie Code supports file uploads

    Uploads are used as chat context, see Attach files for details.

  13. Target an Azure capacity reservation group for classic compute

    Requires VNet injection, guarantees compute capacity for constrained VM types.

    AWS(not on this cloud)Azure, read it on their docsGCP(not on this cloud)SAP(unknown)
  14. Genie Code web search is in Beta

    Searches the public web for current info, disabled by default.

    AWS, read it on their docsAzure(not on this cloud)GCP, read it on their docsSAP(unknown)
  15. Full page Genie Code is now generally available

    Runs multiple chats in parallel, with notebooks and files as tabs.

  16. Serverless compute access control is now generally available

    Governs access to notebooks, jobs, and pipelines through Default Interactive and Automated Compute objects. Existing workloads run unchanged.

  17. Web terminal on serverless GPU compute (AI Runtime) is in Public Preview

    Runs shell commands, monitors GPU usage, and manages files on serverless GPU compute. Requires environment version 5 or above.

    AWS, read it on their docsAzure, read it on their docsGCP(not on this cloud)SAP(unknown)
  18. Dedicated group clusters will soon be available by default for workspaces with the compliance security profile enabled Coming soon

    Enables shared compute for groups with compliance security profile, supporting Databricks Runtime for ML and RDD APIs.

  19. Genie Code scheduled tasks are in Beta

    Runs prompts on a recurring schedule, producing a continuable Genie chat with results.

  20. Get this table by email. One Monday mail covering the week, filtered to the clouds and products you run. Weekly digest
  21. Serverless compute is now available in the Azure UK West region

    Supports notebooks, jobs, pipelines, and SQL warehouses, with private connectivity.

    AWS(not on this cloud)Azure, read it on their docsGCP(not on this cloud)SAP, read it on their docs
  22. Protobuf tensor input for custom model serving endpoints (Public Preview)

    Replaces JSON with serialized KServe v2 ModelInferRequest. Requires endpoints deployed after July 9, 2026.

  23. Genie Code system table now counts only user-submitted interactions

    Excludes background and automated requests, such as autocomplete and quick-fix suggestions.

  24. Genie Code now uses models served through OpenAI on Databricks

    Adds OpenAI to existing Azure OpenAI and Anthropic models.

    AWS, read it on their docsAzure(not on this cloud)GCP, read it on their docsSAP(not on this cloud)
  25. Schedule deferred policy enforcement for all-purpose compute

    Applies on next restart or immediately, and admins can cancel scheduled enforcements.

  26. Instance events and instance pools system tables are now available (Public Preview)

    Track instance state transitions and instance pool configurations via system.compute.instance_events and system.compute.instance_pools tables.

  27. Serve custom LLMs with Custom Model Serving (Beta)

    Supports multimodal models and PEFT recipes, unlike Foundation Model APIs.

    AWS, read it on their docsAzure, read it on their docsGCP(not on this cloud)SAP(not on this cloud)
  28. Databricks Container Services for standard compute is now available (Beta)

    Requires Databricks Runtime 18.3 or later, allows custom Docker images in shared compute.

  29. Alibaba Cloud Qwen3.5 122B A10B now available as a Databricks-hosted model in Public Preview

    Accessible via Foundation Model APIs, available in Public Preview.

    AWS, read it on their docsAzure, read it on their docsGCP(not on this cloud)SAP(not on this cloud)
  30. Compute log delivery to volumes is now GA

    Delivers Spark driver, worker, and event logs to Unity Catalog volumes, recommended for log storage.

  31. Some partner-powered AI features are now available in-country for Brazil, Canada, Japan, Korea, Singapore, and the UK

    Requires Enforce data processing within workspace Geography enabled. Data does not leave the country.

    AWS, read it on their docsAzure(not on this cloud)GCP, read it on their docsSAP(not on this cloud)
  32. Serverless compute is enabled by default

    Eligible Google Cloud workspaces are affected, account admins no longer need to manually enable it.

    AWS(not on this cloud)Azure(not on this cloud)GCP, read it on their docsSAP(not on this cloud)
  33. Endpoint telemetry for custom model serving endpoints (Beta)

    Stores logs, traces, and metrics in Unity Catalog Delta tables using OpenTelemetry.

    AWS, read it on their docsAzure, read it on their docsGCP(not on this cloud)SAP(not on this cloud)
  34. Flexible node types are now generally available

    Falls back to alternative instance types when specified type is unavailable, improving launch reliability.

  35. Function calling and structured output now available using the Alibaba Cloud Qwen3-Next Instruct model (Beta)

    Supports Qwen3-Next 80B A3B Instruct model.

    AWS, read it on their docsAzure(not on this cloud)GCP(not on this cloud)SAP(unknown)
  36. OpenAI GPT 5 models are now generally available in Mosaic AI Model Serving

    Includes GPT-5 mini and nano models.

    AWS, read it on their docsAzure(not on this cloud)GCP, read it on their docsSAP(unknown)
  37. Configure compute to use AWS Capacity Blocks

    Replaces on-demand instances, requires AWS account setup.

    AWS, read it on their docsAzure(not on this cloud)GCP(not on this cloud)SAP(unknown)
  38. Alibaba Cloud Qwen3-Next Instruct now available in Beta

    Supports 80B A3B Instruct in us-west-2 and ap-northeast-1 regions via pay-per-token APIs.

    AWS, read it on their docsAzure, read it on their docsGCP(not on this cloud)SAP(unknown)
  39. Improved autoscaling behavior for Mosaic AI Model Serving

    Ignores brief traffic surges, scaling on sustained load increases. Reduces unnecessary concurrency scaling.

  40. Mosaic AI Model Serving now supports OpenAI GPT 5 models (preview)

    Supports GPT-5, GPT-5 mini, and GPT-5 nano models for batch inference with AI Functions.

    AWS, read it on their docsAzure(not on this cloud)GCP, read it on their docsSAP(unknown)
  41. Anthropic Claude Opus 4.1 now available as a Databricks-hosted foundation model

    Accessible via Foundation Model APIs on a pay-per-token basis.

  42. Route-optimized endpoints now require route-optimized URL path for querying

    Requires route-optimized URL path for querying, standard workspace URL path not supported for new endpoints.

Headlines, dates and product areas are Databricks' own, from the release notes published for each cloud, and every item links to the note it came from. The one-line summaries are ours. Not a Databricks product and not affiliated with Databricks, Inc.