> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://poly-ai.ferndocs.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://poly-ai.ferndocs.com/_mcp/server.

# 1.2025

> January 2025 release notes.

The **January 2025** PolyAI Agent Studio release introduces easier in-browser calling, a new safety dashboard, custom analytics dashboards,
a new Knowledge structure, and updates to functions and the Conversation API.

Expand the items for details:

#### In-app calling

In-app calling is a quicker way to [test voiced conversations](/agent-studio/get-started/quickstart#test-your-agent) with your agent, removing the need for telephony services like [Twilio](/agent-studio/voice/numbers/twilio/overview).

![calling](/_fern-files/_img/3233e33b5428d4c3091d4084bd48ac6506abc264239709e63796c0b697cfd48a.webp)

**How to use:**

* Click the phone icon  in the top-right corner of the project page.
* Select an [agent version](/agent-studio/deployments/overview) in the dropdown to start testing.

#### Audio management

The new **text-to-speech audio cache** detects common phrases across calls, enabling easy customization.

![audio-management](/_fern-files/_img/83cd937130c2e45ce7f5a05008ec09b8fb71503d87df8ae1c4e4fd9e56445144.webp)

**Key Benefits:**

* Enhance voice quality by editing utterances directly in-studio.
* Streamline testing and iteration for commonly used phrases.

Visit the full [audio management](/agent-studio/voice/audio-library) page for details.

#### Knowledge structure

A dedicated **Questions** field has been added to the Knowledge UI, allowing for clearer separation between sample questions and content.

![kb-questions](/_fern-files/_img/2f37129ef91525631149739eb715cff8e954164ab920c270e6bd31a2a9f6d818.webp)

**Key Benefits:**

* Improves readability for builders.
* Simplifies the process of adding and editing sample questions.

Visit the [Managed Topics Introduction](/agent-studio/knowledge/fa-qs/fa-qs) page for more details.

#### Safety dashboard

The enterprise safety dashboard provides insights into flagged calls and safety filter activations.

![safety-dashboard](/_fern-files/_img/3461c9b0d6c93194bb0f0dea9f47cba4638d379570e0405548d20a01c595a6b0.webp)

**Key Benefits:**

* Monitor flagged calls across projects.
* Understand which safety filters were triggered and why.

Visit the full [safety dashboard](/agent-studio/custom-dashboards/safety) page for details.

#### Custom dashboards

Support for custom dashboards is now available for all projects.

![custom-dashboard](/_fern-files/_img/abf4fcbbc9cb40af9ddcede60763e7e713eb924dcaa48c273a05197103c539fc.webp)

**Key Benefits:**

* Tailored dashboards aligned with project-specific goals.
* Improved clarity and focus on critical success metrics.

Visit the full [custom dashboard](/agent-studio/custom-dashboards/custom) page for details.

#### End functions

End functions allow unconditional function calls at the conclusion of conversations.

![end-functions](/_fern-files/_img/b8e832b27aa4ae322eb77d39640bb3acf6f27c70c83ab988cdc8975f584084d2.webp)

**Use Cases:**

* **Data logging:** Save conversation details to a [CRM](/integrations/integrations/overview).
* **Task automation:** Trigger workflows or send follow-up [SMS](/messaging-channel/sms).

Visit the full [End Function](/agent-studio/tools/end-tool) page for implementation details.

#### Variants in the Conversations API

The Conversations API now includes `variant_id` and `variant_name` in the following endpoints:

* [Get Conversations](/api-reference/conversations/v3/endpoint/get-conversations): `/v1/{account_id}/{project_id}/conversations`
* [Get Maximum Concurrent Call Numbers](/api-reference/api-reference/concurrent-calls/get-max): `/v1/{account_id}/{project_id}/conversations/concurrency`

These additions allow for better tracking and management of variant-specific data.