top of page Launches Customizable Large Language Model for Contact Centers

Image credit: Observe.AI

Overview announced the introduction of its Large Language Model (LLM) with a 30-billion-parameter capacity, along with a Generative AI Suite. Aimed at enhancing agent performance, this technology leverages a rich dataset of real-world contact center interactions, setting it apart from models like OpenAI's GPT.'s platform stands out for the calibration and control it offers users, allowing fine-tuning and customization to meet specific contact center requirements. The LLM has undergone specialized training on numerous contact center datasets, equipping it to handle AI-based tasks such as call summarization, automated QA, and coaching.

With the help of LLM's capabilities, aims to improve agent performance across all customer interactions, such as phone calls, chats, queries, and complaints. The company believes this will empower agents to provide superior customer experiences.

Swapnil Jain, CEO of, emphasized the value of quality and relevance in the instruction dataset used in training their LLM. The dataset contained hundreds of instructions across various tasks directly relevant to contact center use cases. This meticulous approach to dataset curation improved the model's ability to provide accurate and contextually appropriate responses.

Initial benchmarks indicate that's LLM outperforms GPT-3.5, demonstrating a 35% boost in conversation summarization accuracy and a 33% improvement in sentiment analysis. The company used redacted data for training, ensuring no personally identifiable information (PII) is used, reflecting its commitment to customer data privacy.

According to Jain, despite their promise, generic LLMs face challenges that limit their effectiveness in contact centers, including lack of specificity and control, and inability to understand human conversation and real-world contexts accurately. addresses these by incorporating five years of well-processed and pertinent data from hundreds of millions of customer interactions.

The company's new generative AI suite empowers agents throughout the entire customer interaction lifecycle, with features such as Knowledge AI for quick and accurate responses and Auto Summary for reducing post-call tasks. The Auto Coaching tool offers immediate, personalized, evidence-based feedback to agents after a customer interaction, supplementing their regular supervisor-based coaching sessions.

Key Takeaways:

  1. launched a 30-billion-parameter Large Language Model (LLM) and a Generative AI Suite designed to enhance contact center agent performance.

  2. The company's LLM outperforms OpenAI's GPT-3.5, showing a 35% improvement in conversation summarization accuracy and a 33% boost in sentiment analysis.

  3. The LLM underwent training exclusively on redacted data, highlighting's commitment to customer data privacy.

  4. The generative AI suite features tools like Knowledge AI, Auto Summary, and Auto Coaching to assist agents throughout the customer interaction lifecycle.

  5. believes generative AI will empower human talent in contact centers to perform with greater efficiency and speed. Source

bottom of page