Yandex AI Studio release notes

Release as of 10/11/2025Release as of 10/11/2025

You can no longer deploy the gpt-oss-20b and gpt-oss-120b models in dedicated instances. These models remain available in the common instance.

Release as of 07/11/2025Release as of 07/11/2025

Added support for Yandex Search API in Yandex Cloud ML SDK.

Release as of 02/10/2025Release as of 02/10/2025

The Llama 8B and Llama 70B models are no longer supported in the common instance.

Release as of 24/09/2025Release as of 24/09/2025

Yandex Foundation Models has evolved into Yandex Cloud AI Studio, a full-value generative model and AI agent platform. There are some further updates:

• Our voice assistants and the Realtime API you need to use them are now available at the Preview stage.
• MCP Hub is now available from the management console. The feature is at the Preview stage.
• Extended compatibility with the OpenAI API. Added support for the Responses API, Realtime API, and Vector Store API.
• Agent Atelier, the agent constructor, is now available from the management console.
• Added the option to manage Vector Store search indexes and search files from the management console.
• The AI agents with the main tools for the Responses API and Vector Store API have now entered the General Availability stage. The AI Assistant API is still there but will not be developed further. Use the Responses API for your new projects.

Release as of 19/09/2025Release as of 19/09/2025

In the management console, users without a billing account will no longer enjoy free requests to YandexGPT and YandexART.

Release as of 16/09/2025Release as of 16/09/2025

Added the ability to deploy some models on dedicated instances.

Release as of 03/09/2025Release as of 03/09/2025

The common instance now features the Gemma 3 27B visual-linguistic model. For text-to-text, access the model in AI Playground; for images, use the OpenAI API or ML SDK.

Release as of 28/08/2025Release as of 28/08/2025

• The YandexGPT Pro 5.1 model is available for testing (RC branch). The new model does not support the reasoning mode.
• Discontinued support for version 4 of the YandexGPT Pro, YandexGPT Pro 32k, YandexGPT Lite models, and models fine-tuned on YandexGPT Lite version 4.

Release as of 06/08/2025Release as of 06/08/2025

The OpenAI gpt-oss-120b and gpt-oss-20b models are now available in synchronous mode via the OpenAI API. These models are good at reasoning, but do not yet support function calling.

Release as of 24/07/2025Release as of 24/07/2025

Updated the Qwen3 235B model by increasing its context length to 256,000 tokens.

Release as of 03/07/2025Release as of 03/07/2025

Qwen3 235B is now available in synchronous mode. The model works only through the OpenAI API.

Release as of 15/05/2025Release as of 15/05/2025

In line with the lifecycle, updated the text generation model versions available in synchronous and asynchronous mode. Discontinued support of YandexGPT version 3 models and models fine-tuned in Yandex DataSphere.

Release as of 30/04/2025Release as of 30/04/2025

New text models of the Qwen3 family are now available in batch mode.

Release as of 24/04/2025Release as of 24/04/2025

Vision language models are now available in AI Studio.

Added the batch mode for working with models: now you can process large amounts of data with a single request. The batch mode is supported for text generation models and vision language models. Added new types of datasets to use in batch mode.

Release as of 31/03/2025Release as of 31/03/2025

1. The RC branch now features the YandexGPT Lite 5th generation model, which supports contexts of up to 32,000 tokens in both synchronous and asynchronous modes.
2. Added support for OpenAI tools to work with text generation models.

Release as of 19/03/2025Release as of 19/03/2025

Increased some limits for AI Assistant API: now you can add up to 10,000 documents with the total size of 5 million tokens to a search index. You can upload up to 100 documents at a time. Also, increased the maximum number of threads to 10,000 and maximum number of messages per thread to 100,000. For a complete list of limits, see Yandex AI Studio quotas and limits.

Release as of 25/02/2025Release as of 25/02/2025

The YandexGPT Pro 5th generation model is now available for testing in the RC branch. The 5th generation key upgrades include:

• Function calling was significantly improved.
• Added support for structured output. This feature enables you to set up the model to generate responses in random JSON format or according to the provided schema. For more information on structuring model output, see Text generation overview.
• Increased the supported context to 32,000 tokens for all modes.

Release as of 11/02/25Release as of 11/02/25

1. Updated the Llama 70B model version. Now Llama 3.3. Llama was created by Meta. Meta is designated as an extremist organization and its activities are prohibited in Russia. is available in all branches.
2. A new version is out: ML SDK 0.3.1. It features the following updates:
   • Python 3.8 is no longer supported.
   • Added the multipart mode for uploading large datasets.
   • Added the allow_data_logging option that allows you to use dataset data to improve the tuning service when loading datasets.
   • Added the validation_errors field that stores dataset validation errors.
   • Replaced the grpc_credentials field with verify in the SDK builder.

Release as of 07/02/25Release as of 07/02/25

Added support for the reasoning mode in the YandexGPT Pro model.

Release as of 09/12/24Release as of 09/12/24

Upon request, LoRA-based model and classifier tuning has been added in Preview.

YandexGPT-based classifiers are now publicly available.

Release as of 04/12/24Release as of 04/12/24

Llama 3.1 models are now available in AI Studio. For model usage costs, see Yandex AI Studio pricing policy.

Release as of 02/12/24Release as of 02/12/24

The YandexGPT 4th generation model is now available in the main branch (Latest). Version 3 will remain available in the Deprecated branch as per the models' lifecycle.

Release as of 21/11/24Release as of 21/11/24

The AI Assistant API functionality is now available to all Yandex AI Studio users at the Preview stage.

Release as of 01/11/24Release as of 01/11/24

1. Image generation with YandexART is now publicly available. Starting November 1, 2024, YandexART is billed as per the rules described on the AI Studio pricing policy page.
2. Increased the YandexART quotas for the number of generation requests per minute and full day (24 hours).
3. Increased the YandexGPT quota for the number of concurrent generations. For information on the restrictions in place, refer to Yandex AI Studio quotas and limits.
4. Starting December 2, 2024, the YandexGPT model's test version (RC branch) will become the main version (Latest branch). The current version will remain available in the Deprecated branch as per the models' lifecycle.

Release as of 24/10/2024Release as of 24/10/2024

1. The YandexGPT 4th generation model is available for testing (RC branch). Compared to the previous generation, the model's response speed has increased by an average of 2.5 times. The maximum context the model operates has also been increased. In asynchronous mode, 4th generation models can process up to 32,000 tokens. And now there is the YandexGPT Pro 32k model added to process large contexts in synchronous mode. For more information on model limitations, see Yandex AI Studio quotas and limits.
2. Increased the maximum number of tokens per response in the management console.

Release as of 10/10/24Release as of 10/10/24

Updated the YandexART model. The updated version has better understanding of prompts, considers more details, and can generate text in Latin characters on the image.

Release as of 21/06/24Release as of 21/06/24

Starting June 24, 2024, the YandexGPT Lite model based on YandexGPT 3 is available by default (latest branch). The deprecated model based on YandexGPT 2 is available in the deprecated branch until July 1, 2024.

Release as of 07/06/24Release as of 07/06/24

Updated the YandexART model:

• Compared to the previous version, the updated model understands text prompts better and creates more realistic images.
• Added the optional aspectRatio parameter for the image aspect ratio.

Release as of 29/05/24Release as of 29/05/24

Added the text classification feature.

Release as of 28/05/24Release as of 28/05/24

The YandexGPT 3-based YandexGPT Lite RC model is now available in Release Candidate status. The model may replace the current YandexGPT Lite in the future.

Release as of 19/04/24Release as of 19/04/24

Now you can send asynchronous requests to YandexGPT models fine-tuned in DataSphere.

Release as of 09/04/24Release as of 09/04/24

1. Added the generation of images based on a text description. The YandexART model works in asynchronous mode and is available in the management console in AI Studio Playground and via the API.
2. Added examples of requests to YandexART in the documentation.

Release as of 25/03/24Release as of 25/03/24

Added a new YandexGPT Pro model of the YandexGPT 3 family.

Release as of 26/01/24Release as of 26/01/24

Updated the YandexGPT and YandexGPT Lite models and improved their response performance.

Release as of 06/12/23Release as of 06/12/23

1. Added a new YandexGPT model for asynchronous mode.
2. Significantly revised the service's API.
3. Unified model names and the way to access the models.

Release as of 02/08/23Release as of 02/08/23

1. Increased the total number of tokens in the prompt and response.
2. Added a new mode called Chat.
3. Added a method for counting the number of tokens in a request.