Product Update

Infographic explaining Balto’s new sentiment model in three steps: seeding a large LLM with high-quality sentiment labels, distilling it into an ~8B model to scale auto-labeling of utterances, and deploying a compact production model that processes ~2,500 requests per second, delivers sentiment scores every ~800 ms, and remains LLM-agnostic.

Balto’s New Sentiment Analysis Model: Moving Beyond Positive and Negative Labels
Chris Kontes Headshot

Chris Kontes

Note: Our customers are increasingly interested in the details behind AI models, so this is a more technical article than usual. From Extremes to Nuance When we first launched Balto’s sentiment analysis over a year ago, we deliberately kept it simple. Calls were labeled as either positive or negative, surfacing conversations at the extremes: the…

Read More
    Infographic explaining Balto’s new sentiment model in three steps: seeding a large LLM with high-quality sentiment labels, distilling it into an ~8B model to scale auto-labeling of utterances, and deploying a compact production model that processes ~2,500 requests per second, delivers sentiment scores every ~800 ms, and remains LLM-agnostic.

    Balto’s New Sentiment Analysis Model: Moving Beyond Positive and Negative Labels

    Chris Kontes Headshot

    Chris Kontes

    Read More