Anthropic Claude Haiku 5.5 is now available on Unity Gateway
Anthropic Claude Haiku 5.5 is now available on Unity Gateway. You can access this model using Foundation Model APIs . Customers are responsible for ensuring their compliance with applicable terms .
Also shipped Oct 7, 2026Databricks in October 2026 →
The microVM sandbox type is now Generally Available (GA) with GKE Sandbox in clusters that run version...
The microVM sandbox type is now Generally Available (GA) with GKE Sandbox in clusters that run version 1.37.0-gke.4713000 and later. MicroVM sandboxes provide hardware virtualization and isolation for untrusted workloads, AI agent runtimes, and multi-tenant environments.
SSH for Cloud Run services and instances is in Preview
SSH for Cloud Run services and instances is in Preview . Use this feature to establish a secure, interactive shell connection to your running instances.
Set credit usage limits and alerts for your workspace
Workspace admins and owners on paid plans can now set usage limits and alerts that keep shared credits under control. Set a credit threshold for the whole workspace, a project, a member, or, on Business and Enterprise plans, a group or an access token, and choose what happens when usage reaches it: an email or in-app alert, or a block that stops building or...
Faster inline text edits
Simple text changes you make with Edit text inline in the preview toolbar now apply in a few seconds. Text that your app builds from data or translates still takes a little longer.
Gemini 3.7 Flash is deprecated for your app's AI features
Google is retiring Gemini 3.7 Flash ( google/gemini-3.7-flash ), so it is now marked deprecated for your app's AI features and stops working on January 28, 2027. Apps that use it keep working until then, and Lovable no longer picks it for new AI features.
The Cost Analysis tab of the Billing page now includes Serverless Inference Live Usage, which lists the...
The Cost Analysis tab of the Billing page now includes Serverless Inference Live Usage, which lists the estimated cost and token counts of your Serverless Inference requests by model and model access key, for both teams and organizations. …
More from Agent Bricks and Mosaic AIEvery Agent Bricks and Mosaic AI release →
Enforce partner guardrails with external service policies (Beta)
You can now enforce a third-party guardrail on Unity Gateway traffic. An external service policy sends request and response content to your guardrail vendor and blocks the call when the vendor denies it, with no changes to your applications. See Enforce a partner guardrail with an external service policy .
Use request tags in Unity Gateway service policies (Beta)
Custom service policies in Unity Gateway can now check caller-supplied request tags, such as requiring a tag before a model request or MCP tool call proceeds. See Request tags .
Use OpenJev as the evaluator for LLM-as-a-judge service policies (Beta)
You can now select OpenJev (Qwen3.5 4B) as the evaluator model service of a custom LLM-as-a-judge service policy. OpenJev classifies content without generating output tokens, so it typically responds faster than a chat evaluator. See Use OpenJev as the evaluator .
Unity Gateway supports Amazon Bedrock Mantle
You can now access Amazon Bedrock Mantle models through Unity Gateway model provider services, with Unity Catalog permissions, rate limits, budgets, and inference tables. See Model providers external to Databricks .
Agent Bricks documentation for code-first agent development
The Agent Bricks documentation now centers on building, deploying, and running custom agents in code. See Develop agents on Databricks : Agent compute stack : New pages describe how a deployed agent's framework or harness, agent server, and agent runtime fit together. See Deploy agents on Databricks .
Unity Gateway Skills API and developer tools are available in Beta
The Unity Gateway Skills REST API is now available in Beta , with create, read, update, list, delete, and finalize support across Databricks developer tools: Databricks CLI Python SDK Go SDK Go Modular SDK Java SDK JavaScript SDK See the Skills API and SDK reference .
Agent memory pipelines are in Beta
Agent memory pipelines are in Beta . A memory pipeline connects a session store to a memory store. Each run uses an LLM to distill durable facts, preferences, and decisions from stored conversations into memory entries that your agent recalls in later conversations. You can run a pipeline on demand or about every 24 hours. See Agent memory pipelines .
Knowledge Assistant and Supervisor Agent are now legacy
Knowledge Assistant and Supervisor Agent are now legacy offerings. For new agents written in code, Databricks recommends the Agent Bricks CLI . To build agents without code, use Genie Agents . See Legacy agent offerings .