Releasesby IbisDatabricks AI Runtime SDK replaces serverless_gpu
Update2 days agoUnverifiedAdded Oct 15, 2026
Starting in GPU environment version 6, the Databricks AI Runtime SDK ( databricks-sdk-air ) replaces the deprecated serverless_gpu package. It provides the @distributed and ray_init APIs and the databricks.air.data utilities under the databricks.air namespace. The gpus and gpu_type arguments are renamed to num_accelerators and accelerator_type .
Topics: Developer tools
More Databricks Data Intelligence Platform releases
Every Databricks Data Intelligence Platform releaseOn-demand state repartitioning is now generally available
On-demand state repartitioning is now generally available. You can resize the number of shuffle partitions for a stateful Structured Streaming query or Lakeflow pipeline without losing checkpoint state. See On-demand state repartitioning for stateful streaming queries .
AI Runtime will soon be available by default for workspaces with the compliance security profile enabled
AI Runtime is scheduled to become generally available on October 15, 2026, on AWS and Azure for workspaces without the compliance security profile enabled. AI Runtime will be available by default for eligible workspaces with the compliance security profile enabled in mid November. Availability depends on the supported compliance standards and regions.
Feature Views will soon be available by default for workspaces with the compliance security profile enabled
Feature Views are now generally available. They will be automatically enabled for workspaces with the compliance security profile enabled in mid-November 2026. Define and compute features declaratively from batch or streaming data, then use them for model training and online serving. See Feature Views .
Feature Views are now generally available
Feature Views are now generally available for batch, streaming, and request-time data. They will be enabled for compliance security profile workspaces in mid-November 2026. See Feature Views .
DeepSeek V4.1 Flash supports priority pay-per-token and reserved provisioned throughput
You can now use DeepSeek V4.1 Flash with priority pay-per-token for latency-sensitive traffic or reserved provisioned throughput for dedicated capacity.
Context-based ingress policies for inbound private connectivity are now generally available
Context-based ingress policies for inbound Private Link are now generally available. Control which registered VPC endpoints can reach your workspaces, and combine endpoint rules with identity and request type conditions in the same policy.
Zhipu AI GLM 5.2 is now a legacy model
GLM 5.2 is now a legacy model. Databricks recommends migrating to GLM 5.3 or GLM 5.3 Flash .
DeepSeek V4 Flash (0731) is scheduled for retirement
DeepSeek V4 Flash (0731) will be retired on November 5, 2026. DeepSeek V4.1 Flash is the recommended replacement. See Deprecated and retired models for migration guidance.
Also shipped on Oct 13, 2026
Databricks in October 2026What's included:
- Three training methods: Supervised fine-tuning (SFT) , direct preference optimization (DPO) , and reinforcement learning with group relative policy optimization (GRPO) . Each method trains a low-rank adapter (LoRA) or the full model weights. Custom loss functions : Train with any objective you compute in PyTorch from the model's log-probabilities.
The AI.AGG function is generally available (GA)
The AI.AGG function is generally available (GA). You can use this function to semantically aggregate unstructured input data based on natural language instructions.
BigQuery generative AI functions now support the gemini-4-argon model
BigQuery generative AI functions now support the gemini-4-argon model. This model is generally available .
Gemini 4 Argon in CodeMender (Preview)
CodeMender now supports Gemini 4 Argon. To use it, pass --model gemini-4-argon to cm find , cm verify , or cm fix . Gemini 3.8 Flash remains the default model. For more information, see Specifying the model .
Gemini 4 Argon and Gemini 4 Argon Cyber Permissive
Gemini 4 Argon ( gemini-4-argon ) is available in General Availability (GA) . Gemini 4 Argon is Google's frontier model that excels at deep work in complex, real-world tasks across domains including coding, finance, legal, cybersecurity, and science.
iOS 27.2 beta 4 (24B5110a)
- View downloads View release notes