Release 10.02.2025

* Compute Cloud: added instructions on how to manage disk encryption via YC CLI.
* Managed Service for Kubernetes: updated k8s versions.
* Object Storage: added a tutorial on how to store Veeam backups in a bucket.
* Smart Web Security: added a tutorial on how to configure Smart protection for backends in Manage Service for Kubernetes cluster via Application Load Balancer Ingress controller.
* Translations updated.
* Fixes and improvements.

Author: DataUI VCS Robot

en/_includes/datasphere/migration/configurations.md

@@ -5,25 +5,25 @@ Since projects run on different {{ yandex-cloud }} platforms with different comp
 {% endnote %}
 
 
-| Name | Number of vCPUs | Number of GPUs | RAM, GB | VRAM, GB |
+Name | Number of vCPUs | Number of GPUs | RAM, GB | VRAM, GB
 ---- | ---- | ---- | ---- | ----
 **Configurations with vCPUs based on [Intel Ice Lake](../../../compute/concepts/performance-levels.md)**
 **c1.4** (default) | 4 | 0 | 32 | —
 **c1.8** | 8 | 0 | 64 | —
 **c1.32** ^1^ | 32 | 0 | 256 | —
 **c1.80** ^2^ | 80 | 0 | 640 | —
-**Configurations with vCPUs based on [Intel Broadwell](../../../compute/concepts/performance-levels.md) and [NVIDIA® Tesla® V100](../../../compute/concepts/gpus.md) GPU**
-**g1.1** | 8 | 1 | From 48 to 96 | 32
-**g1.2** | 16 | 2 | From 96 to 192 | 64
-**g1.4** ^1^ | 32 | 4 | From 192 to 384 | 128
-**Configurations with vCPUs based on [AMD EPYC™](../../../compute/concepts/gpus.md) and [NVIDIA® Ampere® A100](https://www.nvidia.com/en-us/data-center/a100/) GPU**
+**Configurations with [Intel Broadwell](../../../compute/concepts/performance-levels.md) vCPUs and [NVIDIA® Tesla® V100](../../../compute/concepts/gpus.md) GPUs**
+**g1.1** | 8 | 1 | 48 to 96 | 32
+**g1.2** | 16 | 2 | 96 to 192 | 64
+**g1.4** ^1^ | 32 | 4 | 192 to 384 | 128
+**Configurations with [AMD EPYC™](../../../compute/concepts/gpus.md) vCPUs and [NVIDIA® Ampere® A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs**
 **g2.1** ^1^ | 28 | 1 | 119 | 80
 **g2.2** ^1^ | 56 | 2 | 238 | 160
 **g2.4** ^1^ | 112 | 4 | 476 | 320
-**g2.8** ^2^ | 224 | 8 | 952 | From 320 to 640
-**Configurations with vCPUs based on [Intel Ice Lake](../../../compute/concepts/performance-levels.md) and [NVIDIA® Tesla® T4](https://www.nvidia.com/en-us/data-center/tesla-t4/) GPU**
+**g2.8** ^2^ | 224 | 8 | 952 | 320 to 640
+**Configurations with [Intel Ice Lake](../../../compute/concepts/performance-levels.md) vCPUs and [NVIDIA® Tesla® T4](https://www.nvidia.com/en-us/data-center/tesla-t4/) GPUs**
 **gt4.1** ^1^ | 4 | 1 | 16 | 16
-**Configurations with vCPUs based on  [Intel Ice Lake](../../../compute/concepts/performance-levels.md) and [NVIDIA® Tesla® T4i](../../../compute/concepts/gpus.md#t4i) GPU**
+**Configurations with [Intel Ice Lake](../../../compute/concepts/performance-levels.md) vCPUs and [NVIDIA® Tesla® T4i](../../../compute/concepts/gpus.md#t4i) GPUs**
 **gt4i.1** ^1^ | 8 | 1 | 32 | 24 
 
 ^1^ Available after you deposit at least $10 to your billing account or upon request to the support team.

en/_includes/foundation-models/examples/dataset-sdk-cls.md

@@ -0,0 +1,50 @@
+```python
+#!/usr/bin/env python3
+
+from __future__ import annotations
+import asyncio
+import pathlib
+from yandex_cloud_ml_sdk import AsyncYCloudML
+from yandex_cloud_ml_sdk.auth import YandexCloudCLIAuth
+
+
+def local_path(path: str) -> pathlib.Path:
+    return pathlib.Path(__file__).parent / path
+
+
+async def main():
+
+    sdk = AsyncYCloudML(
+        folder_id="<folder_ID>",
+        auth="<API_key>",
+    )
+
+    # Creating a tuning dataset for the {{ gpt-lite }} base model
+    dataset_draft = sdk.datasets.draft_from_path(
+        task_type="<classification_type>",
+        path="<path_to_file>",
+        upload_format="jsonlines",
+        name="multiclass",
+    )
+
+    # Waiting for the data to be uploaded and the dataset to be created
+    operation = await dataset_draft.upload_deferred()
+    dataset = await operation
+    print(f"new {dataset=}")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+Where:
+
+* `<folder_ID>`: [ID of the folder](../../../resource-manager/operations/folder/get-id.md) the [service account](../../../iam/concepts/users/service-accounts.md) was created in.
+* `<API_key>`: Service account [API key](../../../iam/concepts/authorization/api-key.md) you got earlier required for [authentication in the API](../../../foundation-models/api-ref/authentication.md).
+
+    {% include [sdk-auth-details-paragraph](../sdk-auth-details-paragraph.md) %}
+* `<classification_type>`: [Classification type](../../../foundation-models/concepts/classifier/index.md) the model will be tuned for using the new dataset. The possible values are:
+
+    * `TextClassificationMultilabel`: Binary classification or multi-label classification.
+    * `TextClassificationMulticlass`: Multi-class classification.
+* `<file_path>`: Path to the file containing the ready-made examples for model tuning.

en/_includes/foundation-models/examples/dataset-sdk.md

@@ -20,15 +20,15 @@ async def main():
     )
 
     # Creating a tuning dataset for the {{ gpt-lite }} base model
-    dataset_draft = sdk.datasets.from_path_deferred(
+    dataset_draft = sdk.datasets.draft_from_path(
         task_type="TextToTextGeneration",
         path="<path_to_file>",
         upload_format="jsonlines",
         name="YandexGPT tuning",
     )
 
     # Waiting for the data to be uploaded and the dataset to be created
-    operation = await dataset_draft.upload()
+    operation = await dataset_draft.upload_deferred()
     dataset = await operation
     print(f"new {dataset=}")
 

en/_includes/foundation-models/examples/tuning-sdk-cls.md

@@ -0,0 +1,49 @@
+```python
+#!/usr/bin/env python3
+
+from __future__ import annotations
+import pathlib
+import uuid
+from yandex_cloud_ml_sdk import YCloudML
+
+
+def local_path(path: str) -> pathlib.Path:
+    return pathlib.Path(__file__).parent / path
+
+
+def main():
+    sdk = YCloudML(
+        folder_id="<folder_ID>",
+        auth="<API_key>",
+    )
+
+    # Viewing the list of valid datasets
+    for dataset in sdk.datasets.list(status="READY", name_pattern="multiclass"):
+        print(f"List of existing datasets {dataset=}")
+
+    # Setting the tuning dataset and the base model
+    train_dataset = sdk.datasets.get("<dataset_ID>")
+    base_model = sdk.models.text_classifiers("yandexgpt-lite")
+
+    # Defining minimum parameters
+    # To control more parameters, use `base_model.tune_deferred()`
+    tuned_model = base_model.tune(
+        train_dataset, name=str(uuid.uuid4()), classification_type="<classification_type>"
+    )
+    print(f"Resulting {tuned_model}")
+
+    # Starting the tuning
+    classification_result = tuned_model.run("hey!")
+    print(f"{classification_result=}")
+
+    # Saving the URI of the tuned model
+    tuned_uri = tuned_model.uri
+    model = sdk.models.text_classifiers(tuned_uri)
+
+    classification_result = model.run("hey!")
+    print(f"{classification_result=}")
+
+
+if __name__ == "__main__":
+    main()
+```

en/_includes/foundation-models/lora-tuning-preview.md

@@ -0,0 +1 @@
+_Model tuning based on the {{ lora }} method is at the [Preview](../../overview/concepts/launch-stages.md) stage and available upon request. You can fill out the form in the [management console]({{ link-console-main }}/link/foundation-models/)._

en/_includes/foundation-models/public-talks.md

@@ -3,6 +3,10 @@
 
 ## Videos {#video}
 
+### 2025 {#2025}
+
+* {{ yandexart-name }} for businesses. Watch the video on [YouTube](https://www.youtube.com/watch?v=I9Fpsxv8Hlc) or [Yandex](https://runtime.strm.yandex.ru/player/episode/vplefdetiqv7tasxrbyk).
+
 ### 2024 {#2024}
 
 * Creating AI assistants. [Click here to view the video](https://www.youtube.com/watch?v=kKbMaWSi20I).
@@ -37,7 +41,7 @@
 
 ### 2023 {#2023-articles}
 
-* Feedback analysis, chat assistance, and creatives: what {{ yagpt-full-name }}'s skillset. [Read our article](https://vc.ru/services/945084-analiz-otzyvov-pomosh-v-chate-i-kreativy-chto-umeet-yandexgpt-api)
+* Feedback analysis, chat assistance, and creatives: {{ yagpt-full-name }}'s skillset. [Read our article](https://vc.ru/services/945084-analiz-otzyvov-pomosh-v-chate-i-kreativy-chto-umeet-yandexgpt-api)
 
 * How to ditch servers, get your business analytics going, and give {{ yagpt-name }} a test. [Read our article](https://vc.ru/offline/845622-oboitis-bez-serverov-nastroit-biznes-analitiku-i-protestirovat-yandexgpt)
 

en/_includes/foundation-models/tuning-before-beginning.md

@@ -0,0 +1,31 @@
+To use the examples:
+
+{% list tabs group=programming_language %}
+
+- SDK {#sdk}
+
+  1. [Create](../../iam/operations/sa/create.md) a service account and [assign](../../iam/operations/sa/assign-role-for-sa.md) the `ai.editor` [role](../../foundation-models/security/index.md#languageModels-user) to it.
+  1. [Get](../../iam/operations/api-key/create.md) the service account API key and save it.
+
+      {% include [sdk-auth-details-paragraph](./sdk-auth-details-paragraph.md) %}
+  1. Use the [pip](https://pip.pypa.io/en/stable/) package manager to install the [{{ ml-sdk-name }} library](../../foundation-models/sdk/index.md):
+
+      ```bash
+      pip install yandex-cloud-ml-sdk
+      ```
+
+- cURL {#curl}
+
+  1. {% include notitle [ai-before-beginning](./yandexgpt/ai-before-beginning.md) %}
+  1. Install [gRPCurl](https://github.com/fullstorydev/grpcurl).
+  1. {% include [curl](../curl.md) %}
+  1. (Optional) Install the [jq](https://stedolan.github.io/jq/) JSON stream processor.
+  1. [Get an IAM token](../../iam/operations/iam-token/create.md) used for authentication in the API.
+  
+     {% note info %}
+
+     The IAM token has a short [lifetime](../../iam/concepts/authorization/iam-token.md#lifetime): no more than {{ iam-token-lifetime }}.
+
+     {% endnote %}
+   
+{% endlist %}

en/_includes/foundation-models/tuning-dataset-api-step2.md

@@ -0,0 +1,31 @@
+
+Get a link to upload data into the dataset:
+  
+```bash
+grpcurl \
+  -H "Authorization: Bearer <IAM_token>" \
+  -d '{"dataset_id": "<dataset_ID>", "size_bytes": <dataset_size>}' \
+  {{ api-host-llm }}:443 yandex.cloud.ai.dataset.v1.DatasetService/GetUploadDraftUrl | jq
+  ```
+
+Where:
+* `<IAM_token>`: [IAM token](../../iam/concepts/authorization/iam-token.md) of the service account you got [before you started](#before-begin).
+* `<dataset_ID>`: Dataset ID you saved in the previous step.
+* `<dataset_size>`: Size in bytes of the file with data for tuning. In the terminal, you can get the file size using the `ls -l <file_path>` command.
+
+Result:
+
+```text
+{
+  "datasetId": "fdso08c1u1cq********",
+  "uploadUrl": "https://storage.yandexcloud.net/ai-fomo-drafts-prod/b1gt6g8ht345********/fdso08c1u1cq********?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20250120T105352Z&X-Amz-SignedHeaders=content-length%3Bhost&X-Amz-Expires=86400&X-Amz-Credential=YCAJE_WuJJ9D1r6huCoc8I3yO%2F20250120%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Signature=611d7951994ae939acf4d32cc0c154c738d02adb2a04707a704f34ca********"
+}
+```
+
+The `uploadUrl` field of the response contains a link you can use to upload your data into the dataset.
+
+{% note tip %}
+ 
+If you did not use jq, replace all `\u0026` occurrences with `&` in the link to use it to upload the dataset.
+
+{% endnote %}

en/_includes/foundation-models/tuning-dataset-api-step3.md

@@ -0,0 +1,8 @@
+Upload your data by specifying the link you got in the previous step and the path to the fine-tuning data file:
+
+```bash 
+curl \
+  --request PUT \
+  --upload-file <path_to_file> \
+  "<link>"
+```

en/_includes/foundation-models/tuning-dataset-api-step4.md

@@ -0,0 +1,24 @@
+After the data upload is complete, run the dataset validation:
+  
+ ```bash
+ grpcurl \
+   -H "Authorization: Bearer <IAM_token>" \
+   -d '{"dataset_id": "<dataset_ID>"}' \
+   {{ api-host-llm }}:443 yandex.cloud.ai.dataset.v1.DatasetService/Validate
+ ```
+
+Where:
+* `<IAM_token>`: [IAM token](../../iam/concepts/authorization/iam-token.md) of the service account you got [before you started](#before-begin).
+* `<dataset_ID>`: Dataset ID you saved in the previous step.
+
+Result:
+
+```text
+{
+  "id": "fdso01v2jdd4********",
+  "createdAt": "2025-01-20T11:03:48Z",
+  "modifiedAt": "2025-01-20T11:03:48Z"
+}
+```
+
+Save the validation operation ID (`id` field). You will need it in the next step. 

en/_includes/foundation-models/tuning-dataset-api-step5.md

@@ -0,0 +1,30 @@
+Dataset validation may take some time. To find out validation status and get an error report (if any), send this request:
+  
+```bash
+grpcurl \
+  -H "Authorization: Bearer <IAM_token>" \
+  -d '{"operation_id": "<validation_operation_ID>"}' \
+  {{ api-host-llm }}:443 yandex.cloud.operation.OperationService/Get
+```
+
+Where:
+* `<IAM_token>`: [IAM token](../../iam/concepts/authorization/iam-token.md) of the service account you got [before you started](#before-begin).
+* `<validation_operation_ID>`: ID of the validation operation you saved in the previous step.
+
+Result:
+
+```text
+{
+  "id": "fdso01v2jdd4********",
+  "createdAt": "2025-01-20T11:03:48Z",
+  "modifiedAt": "2025-01-20T11:04:46Z",
+  "done": true,
+  "response": {
+    "@type": "type.googleapis.com/yandex.cloud.ai.dataset.v1.ValidateDatasetResponse",
+    "datasetId": "fdso08c1u1cq********",
+    "isValid": true
+  }
+}
+```
+
+The `isValid` field is set to `true`. This means the loaded dataset was validated successfully.

en/_includes/speechkit/async-modes.md

@@ -1,3 +1,3 @@
 For [asynchronous recognition](../../{{ speechkit-slug }}/stt/transcribation.md), a language model that can operate in two modes is available:
 1. In standard mode, recognition is processed in a standard priority queue. This mode works when the `general` model is selected.
-1. In deferred mode, the audio file to be recognized is placed in a low priority queue and processed at the least busy time. [Special pricing](../../{{ speechkit-slug }}/pricing.md#prices-stt) applies to deferred recognition. The time required to process an audio file in deferred mode is 24 hours or less. Recognition in deferred mode is available when the `deferred-general` tag is specified.
+1. In deferred mode, the audio file to be recognized is placed in a low priority queue and processed at the least busy time. [Special pricing](../../{{ speechkit-slug }}/pricing.md#prices-stt) applies to deferred recognition. The time required to process an audio file in deferred mode is 24 hours or less. Recognition in deferred mode is only available with [API v2](../../speechkit/stt/api/transcribation-api.md) when the `deferred-general` tag is specified.

en/_pricing_examples/datasphere/usd-inference.md

@@ -1,7 +1,7 @@
-> 72 × 185,760 × 2 = 26,749,440 units for running a node
+> 72 × 185,760 × 2 = 26,749,440 units for the node
 > 26,749,440 × $0.0000096 = $256.794624
 >
-> Total: $256,794624, cost of using {{ ml-platform-name }}.
+> Total: $256.794624, cost of using {{ ml-platform-name }}.
 
 Where:
 * 72: Number of units per g1.1 configuration.

en/_qa/speechkit/stt.md

@@ -1,5 +1,6 @@
 # Speech recognition (STT)
 
+
 #### Incorrect stress and pronunciation {#stt-stress-pronunciation}
 
 Create a request and attach examples so that developers can make adjustments to the next releases of the speech synthesis model.