Open a page of your app from Chats
You can now ask Lovable in Chats to show you a page in one of your projects. Lovable opens the project's preview at that page, beside the conversation.
You can now ask Lovable in Chats to show you a page in one of your projects. Lovable opens the project's preview at that page, beside the conversation.
Starting in API version 2027-01 , in customer account full-page extensions the Navigation API currententrychange event sets event.from to the page the customer navigated away from, instead of the page they navigated to. This is a breaking change for customer account full-page extensions that read the new page from event.from .
Amazon MSK now supports TLS 1.3 for data in transit in provisioned clusters. This update improves encryption security and helps meet compliance requirements.
InclusionAI's Ling 3.1 Flash 1019 model is now available on AI Gateway, featuring a 1M-token context window for coding and agent workflows.
Vercel is migrating Data Cache to Runtime Cache, which offers increased storage allocations and new usage-based pricing.
The OpenAI GPT-6.1 Terra model is now available for serverless inference on DigitalOcean.
Fast mode for Claude Opus 4.8 is deprecated and will be removed on November 30, 2026. Users should migrate to Claude Opus 5.5 to continue using fast mode.
You can now update the tools in an existing session in the Agents API public beta. Updates replace the full tool list. Function tools and MCP servers change on the next model call; built-in tools change by the next turn. Context compaction now runs in the background by default. Set compaction.async to false for blocking compaction.
OpenAI is implementing expanded violence prevention safeguards and earlier interventions in ChatGPT to improve safety for teen users.
Spanner Graph supports the following enhancements: You can create property graphs from SQL views that call remote functions ( Preview ) to compute node and edge properties. For more information, see Compute properties with remote functions . You can write simplified queries on schemaless dynamic properties without explicit conversions to SQL data types.
Cloudflare Services for Private Origins is generally available. Proxy public hostnames to origins in your private network over Cloudflare WAN, Cloudflare Tunnel, or Cloudflare Mesh. Refer to Private origins and the announcement blog post ↗︎ .
You can now see how current each destination table is in Lakeflow Connect database connector pipelines, in Public Preview . The new Data freshness column in the pipeline's tables list shows the time since the latest source commit that the pipeline ingested for each change data capture (CDC) table.
Markdown and .patch/.diff files opened from Devin's machine now open as a rendered preview, like attachments do, with a toggle back to the source.
Governed mode allows workspace admins to define specific rules for data source access by the Hex agent. Admins can restrict the agent to use only semantic models, specific projects, or curated warehouse tables instead of relying on the Endorsed classification.
Snowflake introduced a public preview of DEFINE SECRET and environment variables for DCM Projects, enabling declarative secret management and parameter masking.
Cortex Training is now available in preview, allowing users to customize open-weight LLMs using a serverless service within Snowflake.
Sigma added treemap charts and dynamic chips to its AI Assistant, and introduced beta API endpoints for swapping data sources.
Users on Organization and Enterprise plans can now publish Weave tools for internal use within their Figma organization.
Explorer provides a unified editor for database management and logs. Users can save investigations as Notebooks to enable reproducible analysis across project teams.
Neon has added support for Google's Gemini 4 Argon model to its AI Gateway, featuring a 1M token context window.
Users can now assign a user-assigned managed identity to individual databases in Azure SQL Database. This enables secure access to files in Azure Blob Storage and Data Lake Storage Gen2 via external tables and OPENROWSET.
Admins can now configure organization-wide limits independently from workspace defaults. This allows for granular control over budgets and model access.
Changes compare the window you asked for with the one before it The change fields on share of voice and sentiment now compare the requested window with the window of the same length just before it.
Following a layout change in ChatGPT source, the ads.advertiser_info object no longer returns image_url. The field has been replaced by image_data, which contains the advertiser image as base64-encoded bytes. If your parser reads ads.advertiser_info.image_url, update it to use ads.advertiser_info.image_data. No changes to request parameters are needed.
You can now host OpenAI gpt-oss-120b ( openai.gpt-oss-120b ) on dedicated AI clusters using the OAI_H100_X4 unit shape in OCI Generative AI. This shape is available in all 11 OC1 commercial regions with local model processing . To use this shape, create a dedicated AI cluster with OAI_H100_X4 , then create an endpoint for the model on that cluster.
We’ve redesigned the Grafana Cloud homepage around what you need when you log in. It brings your recent work, firing alerts and incidents, your on-call shifts, and the state of your observability stack onto one page. The homepage adapts to your stack. On a new stack, it guides you through setting up your first solutions.
In REST Admin API version 2027-01 and later, the line_item_id property on the Gift Card resource is read-only. Shopify ignores the property in create and update requests. Remove it from those requests before you upgrade.
Lovable now remembers the preferences, decisions, and facts you state as you build, and applies them in later conversations without being reminded. Project memory belongs to one project and is shared by everyone who can edit it. Personal memory covers how you want Lovable to work with you, in every project you build.
Auth0 now supports exponential backoff for phone MFA SMS delivery on Enterprise plans to mitigate SMS pumping and toll fraud.
Heroku Sendgrid add-on customers must now create a Sendgrid account to access the service.
The Astro UI now provides a progress card and dedicated logs for Airflow database migrations during runtime upgrades.
Railway now automatically applies security patches for known vulnerabilities and redeploys affected services. Additionally, services are automatically redeployed when database or bucket credentials are rotated.
Braintrust now supports MCP tracing for tool calls and prompt requests in its SDKs. Additionally, SQL queries can now be executed asynchronously to handle large exports and long-running operations.
Prefect 3.9.1 adds custom state discovery, flow run actions, and various bug fixes for task orchestration.
Retool released stable patches 4.34.15 and 4.0.27 containing bug fixes and security patches.
Arize Phoenix v21.0.0 adds a system state table, new annotation kinds for decisions and agents, and runtime cost tracking for models.
This release adds localhost forwarding to the CLI and SDK, introduces pause and resume functionality for sandboxes, and improves credential file security. It also includes updates to prime-traces and prime-runs.
Metabase 65.1 brings maintenance updates across Open Source and Enterprise editions.
Redpanda Connect 4.116.0 adds incremental snapshot mode for MongoDB CDC and cache mode for SFTP, alongside several bug fixes.
The Apache Software Foundation released Apache Kafka 4.4.0, which includes new features and performance improvements.
Snyk now supports scanning Python 3.15 projects via the CLI and SCM integrations.
GPT 6.1 Terra is now available in Retool via OpenAI, enabling its use in AI resource queries and application building.
You can now apply attribute-based access control (ABAC) row filters and column masks to views, materialized views, and streaming tables shared through OpenSharing. Shared views are in Beta and require the ABAC on views preview. See OpenSharing and ABAC .
The password page of a password-protected shared preview link that allows comments now matches Lovable's current design. Password-protected preview links are available on Business and Enterprise plans.
The transcribe-1 model and S1 are deprecated and will be retired on December 31, 2026. Users should migrate to transcribe-1-pro by updating the model header in their API requests.
Dashboard Tile Groups are now in preview, allowing users to organize tiles into collapsible containers and manage them as single units.
Managed OpenTelemetry for GKE is now generally available, allowing users to collect traces, metrics, and logs using a Google-managed collector. It supports OTLP ingestion and improved integration with Cloud Telemetry API.
AlloyDB AI version 1.6.5 adds support for newer Gemini models and the semantic-ranker-512 model while setting gemini-3.5-flash-lite as default.
Horizon Iceberg REST Catalog telemetry is generally available, logging external query engine HTTP API calls to Snowflake account event tables.
Snowflake has released updates for its Data Clean Rooms product.
Amazon SageMaker AI now supports Seekable OCI (SOCI)-indexed container images, allowing training and processing jobs to lazy-load images and start in seconds instead of minutes.
Monte Carlo's REST API v2 is now in public preview, enabling automated onboarding and deployment of collection agents via scripts, Terraform, or AI tools.
Mistral AI released the open weights for Le Chonk, a 1T parameter multimodal model. It is optimized for coding tasks.
Auth0 now supports RFC 9421 HTTP Message Signatures for Tenant Network ACLs, allowing enterprise tenants to enforce cryptographic origin verification and prevent request tampering.
Node.js 26 LTS is now generally available for builds and Vercel Functions, featuring HTTP/2 support by default and the Temporal API.
Mistral Large 4 has been updated with new weights and lower pricing on Vercel AI Gateway, offering improved performance for coding and cybersecurity tasks.
Sail, a Rust-native distributed compute framework, and LakeSail Cloud now connect to Microsoft OneLake. This integration allows teams to run Spark and Python workloads on Fabric using Delta Lake and Apache Iceberg formats without data copies or extra connectors.
Multigres provides automatic failover for Supabase Postgres and is now available in private alpha for organizations on a paid plan.