ドキュメント変更履歴 2026-08-27

--- ahttps://docs.snowflake.com/en/developer-guide/declarative-sharing/tutorials/getting-started+++ bhttps://docs.snowflake.com/en/developer-guide/declarative-sharing/tutorials/getting-started@@ -23,5 +23,5 @@  This tutorial takes Snowflake data providers through the process of creating, publishing, and accessing a %declarative-app%.-The tutorial uses [SnowCLI](/developer-guide/snowflake-cli/index) as well+The tutorial uses [Snowflake CLI](/developer-guide/snowflake-cli/index) as well as a provided workspace directory and a partially complete manifest file to create a %declarative-app-short%. @@ -51,9 +51,9 @@   If you are not familiar with YAML, see [https://yaml.org/spec/](https://yaml.org/spec/). -- You have installed the SnowCLI command-line interface.--  SnowCLI allows you to manage Snowflake objects and perform various tasks.--  For more information on SnowCLI installation, see the [](/developer-guide/snowflake-cli/installation/installation).+- You have installed the Snowflake CLI command-line interface.++  Snowflake CLI allows you to manage Snowflake objects and perform various tasks.++  For more information on Snowflake CLI installation, see the [](/developer-guide/snowflake-cli/installation/installation).  - You'll require access to two Snowflake accounts:@@ -61,5 +61,5 @@     This account should have the necessary privileges to create and manage Snowflake objects     such as databases, schemas, tables, and virtual warehouses.-  - **Consumer account**, A separate test account representing a consumer,+  - **Consumer account**, a separate test account representing a consumer,     used to test the %declarative-app-short% consumer experience.     This account should have the necessary privileges to install apps and access shared data.@@ -87,15 +87,15 @@    archive file, [tutorial-getting-started.zip](/static/samples/sa/tutorial-getting-started.zip)    and save the link/file to your local file system.-2. Unzip the sample files. The tutorial assumes you unpacked files in to the following directories:+2. Unzip the sample files. The tutorial assumes you unpacked files into the following directories:  <blockquote>  - Linux/macOS: `/tmp/tutorial`-- Windows: `C:temp\tutorial`--</blockquote>--For example to unzip the file on Linux/macOS, assuming they were downloaded to your-`Downloads` folder execute the following command:+- Windows: `C:\temp\tutorial`++</blockquote>++For example, to unzip the file on Linux/macOS, assuming they were downloaded to your+`Downloads` folder, execute the following command:  ```bash@@ -108,5 +108,5 @@ - A workspace directory (*app/workspace*) that contains the content shared by the %declarative-app-short%: a notebook (`.ipynb`) file and a readme. - A manifest file that contains the metadata for the %declarative-app-short%, which you will need to make minor modifications to during the tutorial.-- A sample configuration file for SnowCLI, which you can use to configure your Snowflake connection.+- A sample configuration file for Snowflake CLI, which you can use to configure your Snowflake connection.  ### Snowflake CLI configuration@@ -117,5 +117,5 @@ available in [](/developer-guide/snowflake-cli/installation/installation). -After SnowCLI is installed, you need to configure a connection to Snowflake in your+After Snowflake CLI is installed, you need to configure a connection to Snowflake in your [configuration file](#label-cli-config-locations). This tutorial uses `connections.toml`. `connections.toml` can be found in:@@ -124,5 +124,5 @@ - Windows: `%USERPROFILE%\.snowflake\connections.toml`. -For more information on configuring SnowCLI connections, see [](#label-snowcli-define-connections).+For more information on configuring Snowflake CLI connections, see [](#label-snowcli-define-connections).  To add and test a connection:@@ -163,5 +163,5 @@    ``` -8. Optionally, you can use the following to set the default connection for SnowCLI.+8. Optionally, you can use the following to set the default connection for Snowflake CLI.    Setting the default connection allows you to use the connection name without specifying it in every command.    ```snowcli@@ -173,5 +173,5 @@  The `tutorial_connection` connection is used throughout this tutorial to refer to the provider account.-The remainder of this tutorial assumes you have set this as the default connection for SnowCLI.+The remainder of this tutorial assumes you have set this as the default connection for Snowflake CLI.  When connecting to the consumer account, you will use %sf-web-interface%.@@ -182,5 +182,5 @@ ## Create Snowflake objects -During this step, you, as a provider, will use SnowCLI and create the required Snowflake objects.+During this step, you, as a provider, will use Snowflake CLI and create the required Snowflake objects.  - A database (`DB_TO_SHARE`)@@ -191,7 +191,7 @@ At the completion of this tutorial, you'll remove these objects. -Commands are shown at the command line using SnowCLI as well as in combination using files.--For example, the following command creates a database named `DB_TO_SHARE` using SnowCLI and the required connection.+Commands are shown at the command line using Snowflake CLI as well as in combination using files.++For example, the following command creates a database named `DB_TO_SHARE` using Snowflake CLI and the required connection.  <blockquote>@@ -205,5 +205,5 @@ </blockquote> -Assuming the same SQL command exists in a text file named `create_db.sql`, you can also run the command using SnowCLI as follows:+Assuming the same SQL command exists in a text file named `create_db.sql`, you can also run the command using Snowflake CLI as follows:  <blockquote>@@ -556,5 +556,5 @@  Use the provided `4.test-locally-app.sql` file which contains commands to examine the content of a %declarative-app-short%.-In addition the file, `4.test-only-app.sql`, contains commands to test the app but does not attempt to create an database+In addition, the file, `4.test-only-app.sql`, contains commands to test the app but does not attempt to create a database from the application package. @@ -638,5 +638,5 @@ - Can be easily used to expose databases, tables, views, and schemas. - Have a well-defined lifecycle.-- Can be accessed by consumers using [listings](/collaboration/collaboration-listings-about) or app.+- Can be accessed by consumers using [listings](/collaboration/collaboration-listings-about) or an app.  ### Cleanup (Optional)@@ -650,5 +650,5 @@ If the objects you created in this tutorial are no longer needed, you can remove them from the system using the following commands: -As the provider who originally created the objects, run the following commands using SnowCLI+As the provider who originally created the objects, run the following commands using Snowflake CLI  <blockquote> 

--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/postgres/maintenance+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/postgres/maintenance@@ -291,5 +291,5 @@ - Require the PostgreSQL server to drop replication slots, including any used by the connector. - Cannot preserve or migrate replication slots to the new version. See also [](#label-postgres-upgrade-note).-- Require restarting replication of all tables from the snapshot phase.+- Require restarting replication of all tables from the snapshot phase, unless you can stop all writes to the source database for the duration of the upgrade. In that case, you can keep the replicated data and resume with incremental replication only. For more information, see [](#label-postgres-upgrade-incremental-only).  To perform a minor version upgrade, do the following:@@ -302,14 +302,44 @@  1. Remove all tables from replication in the connector by clearing the **Included Table Names** and **Included Table Regex** parameters.-2. Wait until all queues in the connector are empty.-3. Remove the destination tables by dropping or renaming them.-4. Stop the connector, including all Processors and Controller Services.-5. Open the **Incremental Load** group in the connector.-6. Right-click the top Processor in the group, **Read PostgreSQL CDC Stream**, and select **View state**.-7. Click **Clear state**.-8. Click **Close**.-9. Upgrade PostgreSQL.-10. Restart the connector. A new replication slot will be created.-11. Re-add all tables to begin replication.+2. Wait until all queues in the connector are empty, so that every captured change is merged into the destination tables.+3. Stop the connector, including all Processors and Controller Services.+4. Open the **Incremental Load** group in the connector.+5. Clear the state of the CDC processor:+   1. Open the **Incremental Load** group in the connector.+   2. Right-click the top Processor in the group, **Read PostgreSQL CDC Stream**, and select **View state**.+   3. Click **Clear state**.+   4. Click **Close**.+6. Upgrade PostgreSQL.+7. Restart the connector. A new replication slot will be created.+8. Re-add all tables to the **Included Table Names** or **Included Table Regex** parameters.++You don't need to drop or rename the destination tables before you re-add the tables. The connector re-snapshots each table in place, which preserves the destination table object along with dependent objects such as streams. Before it loads the fresh snapshot, the connector saves a copy of the previous contents in an archive table named `<destination_table>_ARCHIVE_<timestamp>`, which you can drop once you've confirmed that the re-snapshot completed. For more information, see [](#label-of-postgres-restart-table-replication).++<a id="label-postgres-upgrade-incremental-only"></a>++### Upgrade without re-snapshotting tables++When an upgrade drops the replication slot, you can avoid re-snapshotting every table if you can stop all writes to the source database while you upgrade. This applies to major version upgrades and to any upgrade to PostgreSQL 17.0 from version 16 or earlier. The connector keeps the data it already replicated and continues with incremental replication only.++No writes of any kind, whether DML or DDL, can reach the replicated database from the moment you stop the connector until all tables are back in incremental replication. The new replication slot starts at the current write-ahead log position, so any change made during that window is lost and there's no way to recover it without a new snapshot.++To upgrade without re-snapshotting tables, do the following:++1. Remove all tables from replication in the connector by clearing the **Included Table Names** and **Included Table Regex** parameters.+2. Wait until all queues in the connector are empty, so that every captured change is merged into the destination tables.+3. Stop the connector, including all Processors and Controller Services.+4. Stop all writes to the source database, and keep them stopped for the rest of this procedure.+5. Clear the state of the CDC processor:+   1. Open the **Incremental Load** group in the connector.+   2. Right-click the top Processor in the group, **Read PostgreSQL CDC Stream**, and select **View state**.+   3. Click **Clear state**.+   4. Click **Close**.+6. Upgrade PostgreSQL.+7. In the `PostgreSQL Ingestion Parameters` context, set the `Ingestion Type` parameter to `incremental`. For more information, see [](/user-guide/data-integration/openflow/connectors/postgres/incremental-replication).+8. Restart the connector. A new replication slot is created.+9. Re-add all tables to the **Included Table Names** or **Included Table Regex** parameters. The tables bypass the snapshot phase and replicate incrementally into the existing destination tables.+10. Confirm that every table reaches the status INCREMENTAL_REPLICATION in the **Table State Store** controller service. For instructions on viewing table state, see [](#label-of-postgres-check-table-replication-status).+11. Resume writes to the source database.+12. Set the `Ingestion Type` parameter back to `full`, so that tables you add later still get a snapshot.  <a id="label-postgres-upgrade-note"></a> 

--- ahttps://docs.snowflake.com/en/user-guide/collaboration/listings/organizational/org-listing-manifest-reference+++ bhttps://docs.snowflake.com/en/user-guide/collaboration/listings/organizational/org-listing-manifest-reference@@ -19,5 +19,5 @@  As a provider, you can use organizational listings to share data products securely within your organization.-A manifest, written in YAML ([https://yaml.org/spec/](https://yaml.org/spec/)) is required to create organization listings programmatically.+A manifest, written in YAML ([https://yaml.org/spec/](https://yaml.org/spec/)), is required to create organization listings programmatically. Use the information provided here to learn about the manifest format and its individual fields. @@ -33,5 +33,5 @@ - Required - Required for organizational listings. -The general format of a organization listing manifest is:+The general format of an organization listing manifest is:  ```yaml@@ -84,5 +84,5 @@ The **optional** `resources` field contains the following name value pairs: -- `resources.documentation` (String, required ): A fully qualified link to a page on your website with more detailed documentation for the listing.+- `resources.documentation` (String, required): A fully qualified link to a page on your website with more detailed documentation for the listing.   Must start with `http` or `https`. - `resources.media` (String, optional): A fully qualified link to an unlisted or public YouTube video for the listing.@@ -152,5 +152,5 @@ For more information see [](#label-listings-data-dictionaries). -### `data_dictionary` example+#### `data_dictionary` example  ```yaml@@ -186,5 +186,5 @@   - title: "Return all weather for the US"     description: "Example of how to select weather information for the United States"-    query: "select * from weather where country_code='USA'";+    query: "select * from weather where country_code='USA'" . . . ```@@ -198,5 +198,5 @@ - `data_attributes.refresh_rate` (required) -  One of the following: Specifies how often your data product is updated in Snowflake.+  Specifies how often your data product is updated in Snowflake. One of the following:    - CONTINUOUSLY@@ -330,5 +330,5 @@  **Required** when `access` isn't specified, but otherwise **optional**.-Defines who can discover the listing. When not present no accounts can discover the listing.+Defines who can discover the listing. When not present, no accounts can discover the listing.  </dd>@@ -342,5 +342,5 @@ </dl> -Both `discovery` and `access`, contain the same child fields.+Both `discovery` and `access` contain the same child fields.  Either:@@ -350,5 +350,5 @@ <dd> -When *true*, all internal accounts can find or access the listing. When *false* no accounts can find or access the listing.+When *true*, all internal accounts can find or access the listing. When *false*, no accounts can find or access the listing.  </dd>@@ -371,5 +371,5 @@ </blockquote> -### `organization_target` examples+### `organization_targets` examples  The following examples show various combinations of the `discovery` and `access` fields.@@ -488,5 +488,5 @@  - `ALL` - All regions can discover or access the listing.-- An array of regions names prefixed with *PUBLIC* which can discover or access the listing.+- An array of region names prefixed with *PUBLIC* which can discover or access the listing.   For example *access_regions: - name: PUBLIC.AWS_US_WEST_2*.   ```yaml@@ -500,5 +500,5 @@ For a complete list of regions, see [](/sql-reference/sql/show-regions). -## `auto_fulfillment`+### `auto_fulfillment`  Cross-Cloud Auto-fulfillment allows the data product associated with a listing@@ -523,10 +523,10 @@      If `refresh_type` is specified as `SUB_DATABASE_WITH_REFERENCE_USAGE`, do not include this setting.-    The refresh schedule for application packages must be defined at the account level and cannot specified at the listing level.+    The refresh schedule for application packages must be defined at the account level and cannot be specified at the listing level.      For more information see [](/collaboration/provider-listings-auto-fulfillment-set-refresh-interval).  - `USING CRON <expression>` - Defines the data product auto-fulfillment refresh schedule.-  The syntax for `USING CRON` and `REPLICATION SCHEDULE` are the same. See [](#label-create-replication-group-parameters).+  The syntax for `USING CRON` and `REPLICATION SCHEDULE` is the same. See [](#label-create-replication-group-parameters).  - `auto_fulfillment.refresh_type` (required when using `auto_fulfillment`): Must be one of -@@ -611,5 +611,5 @@ <a id="label-org-listing-manifest-resharing"></a> -## `resharing`+### `resharing`  The `resharing` field controls whether consumers of the listing can reshare the data product with other accounts.@@ -645,5 +645,5 @@ ``` -The following example enables resharing on a listing to only within consumer organization:+The following example enables resharing on a listing only within the consumer's organization:  ```yaml@@ -657,5 +657,5 @@ ``` -## `custom_attributes`+### `custom_attributes`  The `custom_attributes` field lets providers apply organization-defined custom attributes 

--- ahttps://docs.snowflake.com/en/user-guide/snowflake-cortex/governance-and-availability/ai-cost-management-and-governance+++ bhttps://docs.snowflake.com/en/user-guide/snowflake-cortex/governance-and-availability/ai-cost-management-and-governance@@ -258,4 +258,8 @@ A [shared resource budget](/user-guide/budgets/budget-shared-resources) lets you track and control credit consumption for AI features (such as AI Functions, Cortex Agents, Cortex Code, and %sf-intelligence%) broken down by the team or cost center consuming them. Instead of budgeting a resource that belongs to a single owner or with a single budget, this budget tracks AI features that are used by specific users. Those users are identified with tags, so you can group them into logical units like a cost center or team. For example, if both an engineering team and a finance team call the same AI function, you can set up separate budgets that each track only the credits consumed by their respective tagged users, even though both teams are using the same underlying AI feature. +### Per-user quotas for AI features++[Per-user quotas](/user-guide/budgets/per-user-quotas) let you set monthly and daily credit limits for individual users. Quotas can cover warehouse compute and AI domains, including AI Functions, Cortex Agents, Cortex Code, and %sf-intelligence%. For AI features, you can configure a quota to automatically block AI access for users who reach their limit and release the block when the cycle resets.+ ### Budget capability by feature @@ -275,21 +279,25 @@     <tr>       <td>Cortex Agents</td>-      <td>[Resource budgets](/user-guide/snowflake-cortex/cortex-agents-resource-budgets), [shared resource budgets](/user-guide/budgets/budget-shared-resources)</td>+      <td>[Resource budgets](/user-guide/snowflake-cortex/cortex-agents-resource-budgets), [shared resource budgets](/user-guide/budgets/budget-shared-resources), [per-user quotas](/user-guide/budgets/per-user-quotas)</td>     </tr>     <tr>       <td>Cortex AI Functions</td>-      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources)</td>+      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources), [per-user quotas](/user-guide/budgets/per-user-quotas)</td>     </tr>     <tr>       <td>Cortex Code CLI (Consumption)</td>-      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources), [daily credit usage limits](/user-guide/cortex-code/credit-usage-limit)</td>+      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources), [daily credit usage limits](/user-guide/cortex-code/credit-usage-limit), [per-user quotas](/user-guide/budgets/per-user-quotas)</td>     </tr>     <tr>       <td>Cortex Code in Snowsight</td>-      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources), [daily credit usage limits](/user-guide/cortex-code/credit-usage-limit)</td>+      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources), [daily credit usage limits](/user-guide/cortex-code/credit-usage-limit), [per-user quotas](/user-guide/budgets/per-user-quotas)</td>+    </tr>+    <tr>+      <td>Cortex Code Desktop</td>+      <td>[Shared resource budgets](/user-guide/budgets/budget-shared-resources), [daily credit usage limits](/user-guide/cortex-code/credit-usage-limit), [per-user quotas](/user-guide/budgets/per-user-quotas)</td>     </tr>     <tr>       <td>%sf-intelligence%</td>-      <td>[Resource budgets](/user-guide/snowflake-cortex/snowflake-cowork/cowork-resource-budgets), [shared resource budgets](/user-guide/budgets/budget-shared-resources)</td>+      <td>[Resource budgets](/user-guide/snowflake-cortex/snowflake-cowork/cowork-resource-budgets), [shared resource budgets](/user-guide/budgets/budget-shared-resources), [per-user quotas](/user-guide/budgets/per-user-quotas)</td>     </tr>     <tr>@@ -308,2 +316,4 @@ Budget evaluation and enforcement are calculated periodically rather than instantaneously. After a budget threshold is exceeded, actions can take up to eight hours to take effect under normal operation, or up to two hours when using the latency-optimized option. Budgets are useful for ongoing spend management and policy enforcement, while still being part of a broader cost governance strategy that may also include usage monitoring and internal operational review.++Per-user quota evaluation occurs within minutes. When block enforcement is enabled, Snowflake can automatically deny further AI requests for users who reach their limit. See [Per-user quotas](/user-guide/budgets/per-user-quotas). 

--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-ml/feature-store/monitoring+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-ml/feature-store/monitoring@@ -168,5 +168,5 @@ This tab is only shown when online serving is enabled for at least one feature view in the feature store. -![The Online Serving tab. Three time-series charts are stacked: Total Request count over time, Errors as a percentage, and Latency in milliseconds with p50, p90, and p99 lines. A Feature Group selector dropdown, time range picker, and warehouse selector appear at the top.](/static/images/screens/snowsight-feature-store-online-serving.png)+![The Online Serving tab. Three time-series charts are stacked: Total Request count over time, Errors as a percentage, and Server-side Latencies in milliseconds with p50, p90, and p99 lines. A Scope selector dropdown, time range picker, and warehouse selector appear at the top.](/static/images/screens/snowsight-feature-store-online-serving.png)  <a id="label-fs-monitoring-online-serving-charts"></a>@@ -176,6 +176,5 @@ **Total Request** -The total number of online feature retrieval requests over the selected time window, across all feature groups or for-the selected feature group.+The total number of online feature retrieval requests over the selected time window, across all scopes or for the selected feature group or feature view.  **Errors (%)**@@ -190,9 +189,11 @@ <a id="label-fs-monitoring-filter-feature-group"></a> -### Filter by feature group--Use the **Feature Group** selector at the top of the tab to filter all charts to a specific [feature group](/developer-guide/snowflake-ml/feature-store/feature-groups). Feature-groups are shown in `name$version` format, for example `user_features$v1`. Select **All** to view aggregate metrics-across all feature groups in the feature store.+### Filter by scope++Use the **Scope** selector at the top of the tab to filter all charts. You can view aggregate metrics for the entire feature store or narrow the view to a specific [feature group](/developer-guide/snowflake-ml/feature-store/feature-groups) or feature view:++- **All**: Aggregate metrics across all feature groups and feature views in the feature store.+- **Feature group**: Filter to a specific feature group shown in `name$version` format, for example `user_features$v1`.+- **Feature view**: Filter to a specific feature view shown in `name$version` format.  The Online Serving tab reflects telemetry from the Postgres-backed Online Feature Store. Feature views backed by hybrid@@ -209,5 +210,5 @@ This tab is only shown when at least one stream feature view exists in the feature store. -![The Stream Ingest tab showing four time-series charts: Ingest API responses per second, Ingest API responses by HTTP code, Ingest API latency with p50, p90, p95, and p99 lines, and Concurrency showing concurrent request volume over time.](/static/images/screens/snowsight-feature-store-stream-ingest.png)+![The Stream Ingest tab showing four time-series charts: API Requests count over time, API Requests by HTTP Code count, API Latency in milliseconds with p50, p90, and p99 lines, and Concurrency in requests per node over time.](/static/images/screens/snowsight-feature-store-stream-ingest.png)  <a id="label-fs-monitoring-stream-ingest-charts"></a> 

--- ahttps://docs.snowflake.com/en/developer-guide/native-apps/container-compute-pool+++ bhttps://docs.snowflake.com/en/developer-guide/native-apps/container-compute-pool@@ -18,5 +18,5 @@ - [](/developer-guide/native-apps/native-apps-about) -The topic describes how to use compute pools in a %na-spcs-plural%.+This topic describes how to use compute pools in a %na-spcs-plural%.  ## About compute pools in apps with containers@@ -24,5 +24,5 @@ A [compute pool](/developer-guide/snowpark-container-services/working-with-compute-pool) is a collection of one or more virtual machine (VM) nodes on which Snowflake runs Snowpark Container Services.-Apps with containers uses a compute pool in the consumer account to manage the container images required by+Apps with containers use a compute pool in the consumer account to manage the container images required by the app. @@ -42,5 +42,5 @@ ## Best practices for using compute pools in an app with containers -Providers should consider the following best practices when creating compute pools a consumer account:+Providers should consider the following best practices when creating compute pools in a consumer account:  - Compute pools have cost implications. It is important to set values for the `min_nodes`,@@ -85,5 +85,5 @@ ## Set the `uses_gpu` property in the manifest file -If an app with containers specifies a GPU as the instance family for the Compute Pool, providers+If an app with containers specifies a GPU as the instance family for the compute pool, providers must set the `uses_gpu` flag to `true` in the manifest. The following example shows how to set this flag in the `artifacts` block:@@ -118,5 +118,5 @@ ### Request the CREATE COMPUTE POOL privilege -An app can request the CREATE COMPUTE POOL privileges from a consumer. This privilege allows the app+An app can request the CREATE COMPUTE POOL privilege from a consumer. This privilege allows the app to create a compute pool in the consumer account. See [Request global privileges from consumers](/developer-guide/native-apps/requesting-privs)@@ -219,5 +219,5 @@  To drop an app with containers that creates a compute pool or warehouse, the-consumer must use the `CASCADE` option to drop the app-owned resources. Attempt to drop+consumer must use the `CASCADE` option to drop the app-owned resources. An attempt to drop without the cascade option will fail.  

--- ahttps://docs.snowflake.com/en/sql-reference/sql/create-dynamic-table+++ bhttps://docs.snowflake.com/en/sql-reference/sql/create-dynamic-table@@ -475,16 +475,4 @@ <dt>`ADAPTIVE`</dt> <dd>--<div className="previewfeat sidebar">--<div className="sidebar-title">--%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open--</div>--Available to all accounts.--</div>  Uses incremental refresh by default but automatically reinitializes the dynamic table when internal heuristics 

--- ahttps://docs.snowflake.com/en/user-guide/collaboration/listings/organizational/org-listing-create+++ bhttps://docs.snowflake.com/en/user-guide/collaboration/listings/organizational/org-listing-create@@ -90,5 +90,5 @@ | Privilege                             | Object   | Notes                                              | | ------------------------------------- | -------- | -------------------------------------------------- |-| CREATE SHARE                          | ACCOUNT  | To *CREATE* a share                                |+| CREATE SHARE                          | ACCOUNT  | To *CREATE* a share.                               | | OWNERSHIP or USAGE with grants option | DATABASE | To see and *USE* the specified database.           | | OWNERSHIP or USAGE with grants option | SCHEMA   | To see the specified schema.                       |@@ -156,5 +156,5 @@ | --------------------------- | ------- | --------------------------------------------------- | | IMPORT ORGANIZATION LISTING | ACCOUNT | To import an organizational listing.                |-| CREATE database             | ACCOUNT | To create a database and mount the listing objects. |+| CREATE DATABASE             | ACCOUNT | To create a database and mount the listing objects. |  </div>@@ -292,5 +292,5 @@ 1. Complete the listing. -   Enter addition information about listing page to guide consumers,+   Enter additional information about the listing page to guide consumers,    such as the following. For more information about these fields, see [](/collaboration/provider-listings-reference). @@ -311,5 +311,5 @@ Create an organizational listing from the share with the required attributes included in YAML (entered in $$ delimiters). -This part of the manifest yaml specifies the accounts that will be able to use the organizational listing:+This part of the manifest YAML specifies the accounts that will be able to use the organizational listing:  ```yaml 

--- ahttps://docs.snowflake.com/en/developer-guide/declarative-sharing/consumer/install+++ bhttps://docs.snowflake.com/en/developer-guide/declarative-sharing/consumer/install@@ -34,5 +34,5 @@ ## Prerequisites -To install a %declarative-app-short%, you must have a Snowflake account, and a role with either of the following privileges:+To install a %declarative-app-short%, you must have a Snowflake account and a role with either of the following:  - The **ACCOUNTADMIN** role@@ -45,5 +45,5 @@ An ACCOUNTADMIN can allow members of the organization to install %declarative-app-short-plural% by granting privileges to the member's role,-using the [GRANT privileges TO ROLE](/sql-reference/sql/grant-privilege) commands:+using the [GRANT privileges TO ROLE](/sql-reference/sql/grant-privilege) command:  ```sqlsyntax@@ -115,5 +115,5 @@ App owners can share access to all of the data and features in an app with the command: [GRANT IMPORTED PRIVILEGES ON APPLICATION](/sql-reference/sql/grant-privilege). -In this example, an app owner imports privileges for the application: `marketing_data_app` to the `team_admin_role` organizational role:+In this example, an app owner imports privileges for the application `marketing_data_app` to the `team_admin_role` organizational role:  ```sql@@ -137,5 +137,5 @@    The command lists the available app roles. If the app has no app roles, the command returns an empty result set. -2. Grant app roles to teams by their organization roles with the command: [GRANT APPLICATION ROLE ...TO ROLE](/sql-reference/sql/grant-application-role) command.+2. Grant app roles to teams by their organization roles with the command: [GRANT APPLICATION ROLE ... TO ROLE](/sql-reference/sql/grant-application-role).    ```sql    GRANT APPLICATION ROLE marketing_data_app.sales TO ROLE sales_team_west; 

--- ahttps://docs.snowflake.com/en/release-notes/preview-features+++ bhttps://docs.snowflake.com/en/release-notes/preview-features@@ -78,4 +78,11 @@   </thead>   <tbody>+    <tr>+      <td>Remote Development with the %sf-vscext%</td>+      <td>Open</td>+      <td>August 2026</td>+      <td>[](/user-guide/vscode-ext-remote-development)</td>+      <td>Connect to a Snowflake-backed remote development environment from Visual Studio Code, Cursor, or Cortex Code Desktop to run notebooks, SQL, scripts, terminals, and Cortex Code on Snowflake-managed compute.</td>+    </tr>     <tr>       <td>Request Access in Workspaces</td> 

--- ahttps://docs.snowflake.com/en/developer-guide/streamlit/app-development/runtime-environments+++ bhttps://docs.snowflake.com/en/developer-guide/streamlit/app-development/runtime-environments@@ -161,5 +161,10 @@  When you use a container runtime, you must select a compute pool to run the Streamlit app.-Each Streamlit app runs on a single compute pool node; a Streamlit app takes an entire node.++When you select `SYSTEM_COMPUTE_POOL_CPU` as the compute pool for your+Streamlit app, Snowflake runs three apps per compute pool node.+Otherwise, each Streamlit app runs on a single compute pool node; a Streamlit+app takes an entire node.+ The size of the compute pool node affects the performance of the app. Larger node sizes can be used if your app requires more memory. However, because Streamlit runs as a single process, 

--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-cli/command-reference/dcm-commands/describe+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-cli/command-reference/dcm-commands/describe@@ -6,5 +6,5 @@ - [](/user-guide/dcm-projects/dcm-projects-overview) -Provides description of a DCM Project.+Provides a description of a DCM Project.  ## Syntax@@ -59,5 +59,5 @@ <dd> -Identifier of DCM Project. Example: MY_DB.MY_SCHEMA.MY_PROJECT. Supports fully qualified (recommended) or simple names. If unqualified, it defaults to the connection's database and schema. Optional if *–target* or *default_target* is defined in the manifest.+Identifier of DCM Project. Example: MY_DB.MY_SCHEMA.MY_PROJECT. Supports fully qualified (recommended) or simple names. If unqualified, it defaults to the connection's database and schema. Optional if *--target* or *default_target* is defined in the manifest.  </dd>@@ -259,5 +259,5 @@  </dd>-<dt>`--format [TABLE%JSON%JSON_EXT|CSV]`</dt>+<dt>`--format [TABLE|JSON|JSON_EXT|CSV]`</dt> <dd>  

--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-cli/command-reference/dcm-commands/test+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-cli/command-reference/dcm-commands/test@@ -9,5 +9,5 @@ Available to all accounts. -Tests all expectations defined in DCM project. It applies only to deployed objects.+Tests all expectations defined in the DCM project. It applies only to deployed objects.  ## Syntax@@ -63,5 +63,5 @@ <dd> -Identifier of DCM Project. Example: MY_DB.MY_SCHEMA.MY_PROJECT. Supports fully qualified (recommended) or simple names. If unqualified, it defaults to the connection's database and schema. Optional if *–target* or *default_target* is defined in the manifest.+Identifier of the DCM project. Example: MY_DB.MY_SCHEMA.MY_PROJECT. Supports fully qualified (recommended) or simple names. If unqualified, it defaults to the connection's database and schema. Optional if *--target* or *default_target* is defined in the manifest.  </dd>@@ -269,5 +269,5 @@  </dd>-<dt>`--format [TABLE%JSON%JSON_EXT|CSV]`</dt>+<dt>`--format [TABLE|JSON|JSON_EXT|CSV]`</dt> <dd>  

--- ahttps://docs.snowflake.com/en/developer-guide/udf/python/udf-python-tabular-functions+++ bhttps://docs.snowflake.com/en/developer-guide/udf/python/udf-python-tabular-functions@@ -46,7 +46,4 @@ - 3.12 - 3.13--Preview versions:- - 3.14 @@ -583,7 +580,4 @@ - 3.12 - 3.13--Preview versions:- - 3.14  

--- ahttps://docs.snowflake.com/en/user-guide/ui-snowsight/notebooks-in-workspaces/notebooks-in-workspaces-compute-setup+++ bhttps://docs.snowflake.com/en/user-guide/ui-snowsight/notebooks-in-workspaces/notebooks-in-workspaces-compute-setup@@ -226,10 +226,10 @@  After a notebook service enters the `RUNNING` state (whether newly created or resumed after being in `SUSPENDED` state), it is guaranteed-not to be disrupted for seven calendar days (168 hours) due to service maintenance. After seven days of creation, the service may be suspended for mandatory+not to be disrupted for seven calendar days (168 hours) due to service maintenance. Seven days after creation, the service may be suspended for mandatory maintenance. -## Multi-node Distributed Training Support (with Snowflake Container Runtime 2.3 or above)+## Multi-node distributed training support (with Snowflake Container Runtime 2.3 or above)  This notebook is optimized for Snowflake Container Runtime 2.3 or above, which introduces support for multi-node clusters. This allows you to scale your ML workloads (like PyTorch, XGBoost, and LightGBM) across multiple nodes for faster training. -Use the doc link to get more information on running ML workloads on multi-node clusters, refer to ths documentation: [https://docs.snowflake.com/en/developer-guide/snowflake-ml/container-runtime-multi-node](https://docs.snowflake.com/en/developer-guide/snowflake-ml/container-runtime-multi-node)+For more information on running ML workloads on multi-node clusters, see [](/developer-guide/snowflake-ml/container-runtime-multi-node). 

--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-cli/command-reference/dcm-commands/drop-deployment+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-cli/command-reference/dcm-commands/drop-deployment@@ -61,5 +61,5 @@ <dd> -Identifier of DCM Project. Example: MY_DB.MY_SCHEMA.MY_PROJECT. Supports fully qualified (recommended) or simple names. If unqualified, it defaults to the connection's database and schema. Optional if *–target* or *default_target* is defined in the manifest.+Identifier of DCM Project. Example: MY_DB.MY_SCHEMA.MY_PROJECT. Supports fully qualified (recommended) or simple names. If unqualified, it defaults to the connection's database and schema. Optional if *--target* or *default_target* is defined in the manifest.  </dd>@@ -72,5 +72,5 @@ <dd> -Name or alias of the deployment to drop. For names containing '$', use single quotes to prevent shell expansion (e.g., 'DEPLOYMENT$1'). If both the deployment name and the alias match two different deployments, the deployment name match has higher precedence.+Name or alias of the deployment to drop. For names containing '$', use single quotes to prevent shell expansion (for example, 'DEPLOYMENT$1'). If both the deployment name and the alias match two different deployments, the deployment name match has higher precedence.  </dd>@@ -273,5 +273,5 @@  </dd>-<dt>`--format [TABLE%JSON%JSON_EXT|CSV]`</dt>+<dt>`--format [TABLE|JSON|JSON_EXT|CSV]`</dt> <dd>  

--- ahttps://docs.snowflake.com/en/developer-guide/streamlit/object-management/billing+++ bhttps://docs.snowflake.com/en/developer-guide/streamlit/object-management/billing@@ -41,5 +41,8 @@ service. The Streamlit server runs continuously on a node of the compute pool, allowing viewers to quickly access the app. Concurrent viewers connect to a single Streamlit-server. Only a single app can run on a node in the compute pool; a Streamlit app+server.++When an app runs on `SYSTEM_COMPUTE_POOL_CPU`, Snowflake runs three Streamlit apps per+compute pool node. Otherwise, a single app can run on a node in the compute pool; a Streamlit app takes an entire node.  

--- ahttps://docs.snowflake.com/en/user-guide/cortex-code/cortex-code-desktop/onboarding-and-authentication+++ bhttps://docs.snowflake.com/en/user-guide/cortex-code/cortex-code-desktop/onboarding-and-authentication@@ -7,4 +7,9 @@  **[Download the latest version](https://www.snowflake.com/en/product/snowflake-coco/downloads/)**. Available for macOS and Windows.++**Windows installer choice:** The downloads page offers two Windows installer types:++- **User installer (recommended)**: Installs CoCo Desktop for the current user only. Doesn't require administrator rights and supports automatic updates without needing elevated permissions.+- **System installer**: Installs for all users on the machine. Requires administrator rights and is intended for IT-managed deployments to managed workstations or when you want a machine-wide install.  Before you sign in, confirm your Snowflake role has the required database roles: 

--- ahttps://docs.snowflake.com/en/user-guide/snowflake-cortex/snowflake-cowork/cowork-resource-budgets+++ bhttps://docs.snowflake.com/en/user-guide/snowflake-cortex/snowflake-cowork/cowork-resource-budgets@@ -4,5 +4,7 @@  - [](/user-guide/budgets)+- [](/user-guide/budgets/per-user-quotas) - [](/user-guide/snowflake-cortex/snowflake-cowork)+- [](/user-guide/snowflake-cortex/snowflake-cowork/cowork-shared-budgets) - [](/user-guide/object-tagging/introduction) @@ -11,4 +13,6 @@ exceeds your configured limits. Resource budgets give you control over the credits consumed at an aggregated level by the entire %sf-intelligence% service.++Resource budgets apply to the %sf-intelligence% object. To set the same monthly or daily credit limit for each user, and optionally block users who reach that limit on %sf-intelligence% and other AI domains, use [per-user quotas](/user-guide/budgets/per-user-quotas).  ## How resource budgets work 

--- ahttps://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents-shared-budgets+++ bhttps://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents-shared-budgets@@ -3,5 +3,7 @@ - [](/user-guide/snowflake-cortex/cortex-agents) - [](/user-guide/snowflake-cortex/cortex-agents-manage)+- [](/user-guide/snowflake-cortex/cortex-agents-resource-budgets) - [](/user-guide/budgets/budget-shared-resources)+- [](/user-guide/budgets/per-user-quotas)  A shared resource-level budget lets you set a credit spending limit for a subset of users on a %cortex-agent%@@ -16,4 +18,6 @@ subject to multiple budgets, each budget is evaluated independently and the user is stopped by whichever threshold is reached first.++Shared resource-level budgets set a spending limit for a tagged group of users on a specific agent. To apply the same monthly or daily credit limit to each user, across Cortex Agents and other AI domains, and optionally block users who reach that limit, use [per-user quotas](/user-guide/budgets/per-user-quotas).  ## How shared resource-level budgets work 

--- ahttps://docs.snowflake.com/en/user-guide/snowflake-cortex/snowflake-cowork+++ bhttps://docs.snowflake.com/en/user-guide/snowflake-cortex/snowflake-cowork@@ -159,4 +159,8 @@    the final answer, including summaries, tables, or charts, in the %sf-intelligence% UI. +## Cost considerations++To monitor spend for a %sf-intelligence% object and act when that spend crosses a threshold, use [resource budgets](/user-guide/snowflake-cortex/snowflake-cowork/cowork-resource-budgets). To enforce monthly or daily credit limits for individual users, and optionally block users who reach their AI quota, use [per-user quotas](/user-guide/budgets/per-user-quotas).+ ## Legal notices  

--- ahttps://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents-resource-budgets+++ bhttps://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents-resource-budgets@@ -4,5 +4,7 @@  - [](/user-guide/budgets)+- [](/user-guide/budgets/per-user-quotas) - [](/user-guide/snowflake-cortex/cortex-agents)+- [](/user-guide/snowflake-cortex/cortex-agents-shared-budgets) - [](/user-guide/object-tagging/introduction) @@ -10,4 +12,6 @@ to control costs for Cortex Agents and take automated actions such as revoking access when spending exceeds your configured limits. Resource budgets give you control over the credits consumed at an aggregated level for that specific Agent.++Resource budgets apply to the Cortex Agent object. To set the same monthly or daily credit limit for each user, and optionally block users who reach that limit on Cortex Agents and other AI domains, use [per-user quotas](/user-guide/budgets/per-user-quotas).  ## How resource budgets work 

--- ahttps://docs.snowflake.com/en/user-guide/snowpipe-streaming/snowpipe-streaming-classic-configuration+++ bhttps://docs.snowflake.com/en/user-guide/snowpipe-streaming/snowpipe-streaming-classic-configuration@@ -28,5 +28,5 @@     You can provide an unencrypted key, or you can provide an encrypted key and provide the `snowflake.private.key.passphrase` parameter to enable Snowflake to decrypt the key. Use this parameter *if and only if* the `snowflake.private.key` parameter value is encrypted. -- `OAuth` : Snowflake OAuth. This option is only available with Snowflake Ingest SDK versions 2.0.3 and later. Configure the following parameters for Snowflake OAuth in the `profile.json` file:+- `OAuth` : Snowflake OAuth. This option is only available with the Java Snowflake Ingest SDK (versions 2.0.3 and later). Configure the following parameters for Snowflake OAuth in the `profile.json` file:    - `oauth_client_id` : The client ID of the OAuth integration.@@ -88,5 +88,5 @@ ### Using Snowflake OAuth -With Snowflake Ingest SDK versions 2.0.3 and later, or Snowflake Connector for Kafka versions 2.1.2 and later, you can use Snowflake OAuth as an authorization method.+With the Java Snowflake Ingest SDK (versions 2.0.3 and later) or Snowflake Connector for Kafka (versions 2.1.2 and later), you can use Snowflake OAuth as an authorization method.  Follow [the workflow](/user-guide/oauth-custom) to create a Snowflake OAuth integration and to call OAuth endpoints to request authorization codes and refresh access tokens. The response of token requests contains `oauth_refresh_token`. After a Snowflake OAuth integration is created, run the [](/sql-reference/functions/system_show_oauth_client_secrets) function to obtain `oauth_client_id` and `oauth_client_secret`. 

--- ahttps://docs.snowflake.com/en/user-guide/budgets/per-user-quotas+++ bhttps://docs.snowflake.com/en/user-guide/budgets/per-user-quotas@@ -4,4 +4,8 @@ - [](/user-guide/budgets/custom-budget) - [](/user-guide/cost-access-control)+- [](/user-guide/snowflake-cortex/cortex-agents)+- [](/user-guide/snowflake-cortex/cortex-agents-resource-budgets)+- [](/user-guide/snowflake-cortex/snowflake-cowork)+- [](/user-guide/snowflake-cortex/snowflake-cowork/cowork-resource-budgets)  Per-user quotas let you enforce spending limits at a per-user granularity, preventing individual 

--- ahttps://docs.snowflake.com/en/sql-reference/sql/create-interactive-warehouse+++ bhttps://docs.snowflake.com/en/sql-reference/sql/create-interactive-warehouse@@ -276,5 +276,7 @@ Specifies the time, in seconds, after which a running SQL statement is canceled. Interactive warehouses have a maximum timeout interval of five seconds.-Any larger values are ignored.+If you specify a value greater than five seconds, Snowflake caps the timeout at five seconds.++For example, if you set `STATEMENT_TIMEOUT_IN_SECONDS = 30`, Snowflake uses the value `5`.  </dd>