Hello everyone, 

We are enhancing Zia's Generative AI service to make your support experience smarter. Here's how:

One of the key challenges in AI is delivering responses that are both contextually accurate and empathetic while being able to consolidate ticket and thread conversations to generate meaningful summaries. The introduction of Qwen into Zia’s Generative AI service will make all of this easier for your agents to achieve while dealing with real customer issues.

While the Zia Generative AI model Llama 3.1 is an 8 billion parameter model known for its faster response generation, Qwen 2.5 is a 32 billion parameter model which trains on a larger scale of data to provide better contextual output. 

Qwen provides enhanced accuracy in generating content and ticket summarization, which enhances the overall performance and reliability of Generative AI features in Zoho Desk. 

Support agents are always subject to time pressure when they're working with users who are facing an issue while using your product or service. Each ticket captures the conversation, from back-and-forth questions, clarifications, and discussions that eventually lead to the solution. This makes tickets a valuable repository of real-world problems and solutions.

However, tickets are not always accessible to everyone. As a result, this knowledge often remains non-reusable. Allowing agents to convert tickets into knowledge base (KB) articles ensures that these insights are captured, shared, and reused effectively.

Agents can select 1–10 conversations, including incoming threads, public and private comments, and agent replies, and convert them into an article that summarizes the problem and solution.

The generated article will be standardized and not include any personal or customer-specific details. Also, the agent cannot select attachments or images shared within the ticket while generating the article.

Article generation is now available for users in the EU, AU, US, CA, JP, and IN data centers. Qwen is available for users across EU, AU, US, and IN data centers. Kindly give the features a try and let us know your feedback in the comments.

For more information on article generation and Qwen, please refer to the help docs. 

PM: @Kavil Rawat