DigestAI news desk

Cut through the AI noise.

Enterprise & Industry3 min read

Salesforce adds Koa Decisions with image and audio support to workflows

Salesforce launched Koa Decisions on November 6, 2026, a decision model built on Google’s open-weight Gemma that processes text, images, and audio. It replaces HyperClassifier, the predecessor embedded in Agentforce since Spring 2026, and aims to expand into Salesforce workflows and developer tools.

1 source

Key points

  • Koa Decisions processes text, images, and audio, built on Google’s open-weight Gemma model
  • Salesforce claims **85.5% accuracy** on intent-routing benchmarks with **half-second responses**
  • Demo available now; enterprise integrations (flows, Apex, API) pending customer agreements

Koa Decisions uses a single-token output to classify requests, score decisions, or return yes/no answers in milliseconds. Salesforce claims it achieves 85.5% accuracy on a production intent-routing benchmark, with responses in about half a second—up to 7.5 times faster than frontier models under identical conditions. Unlike traditional LLMs, which generate text step-by-step, Koa Decisions emits compact, actionable outputs, reducing compute costs and enabling real-time automation. The demo, available via a free Trailhead account, lets users test category-based routing without retraining, contrasting with enterprise projects that historically took months to deploy.

Full story from Unite.AI · by Aiden Cross, AI Product Strategy & Execution, AI Research AgentOpen source ↗

Salesforce Introduces Koa Decisions with Image and Audio Support

Unite.AI · 6 November 2026

Loading the full article…

This text was published by Unite.AI and written by Aiden Cross, AI Product Strategy & Execution, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page
SalesforceGoogleNVIDIAGemmaKoa DecisionsHyperClassifierJev/TypeSafe

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Enterprise & Industry

All →

Related stories