Decagon Introduces Alto, a Model That Decides When Its Voice Agent Speaks and What It Says
Decagon introduced Alto on November 11, 2026, a Decagon Labs model that controls the conversational layer of its duplex voice agent. The company reports a resolution rate gain of more than five percentage points on live customer calls, with average responses in about 700 milliseconds.
Decagon introduced Alto on November 11, 2026, a Decagon Labs model that controls the conversational layer of its duplex voice agent. Alto decides when the agent speaks and what it says.
Reported results
According to Decagon, Alto lifted resolution rates by more than five percentage points on live customer calls. The company also reports that the model delivers average responses in about 700 milliseconds.
Decagon said the rest of the agent's logic remained effectively unchanged during the resolution comparison. The company presents the resolution gain as coming from the conversational layer.
These figures are Decagon's own claims, as reported by Unite.AI. Benchmark methodology, call volumes and the baseline used in the comparison are not part of the summary available at the time of writing.
What Alto does
Alto sits in the conversational layer of Decagon's duplex voice agent. Its two jobs, per Decagon, are turn-taking (deciding when to speak) and response content (deciding what to say).
Parameter count, context window, pricing and training cutoff are not yet disclosed.
What this means
Decagon is positioning conversation control as a separate component from the agent's other logic. Its reported comparison held that other logic effectively constant, so the resolution gain is attributed to the conversational layer alone. The figures come from live customer calls rather than a lab benchmark, but they are company-reported and have not been independently verified.
Related Articles
Google confirms Gemini app with Gemini Live is coming to Apple CarPlay, rolling out globally from today
Google has confirmed that the Gemini app is coming to Apple CarPlay, bringing the full Gemini Live conversation experience to iPhone users in the car. The rollout is global and starts today for all Gemini users on iOS, according to Google, as reported by 9to5Mac.
Radical Numerics unveils Omnii Defense, an offline model that scores DNA sequences for pathogen threat
Radical Numerics announced Omnii Defense, an edge model that flags natural and AI-engineered pathogens from a laptop and a USB gene sequencer, with no internet connection required. Access is limited to vetted external testers while national laboratories, government agencies and independent evaluators assess it.
ElevenLabs previews Interaction Engine, Memory, Routines and ElevenAI at its New York Summit
At its New York Summit, ElevenLabs previewed Interaction Engine, an orchestration system for conversational agents, and announced Memory and Routines for ElevenAgents. It also introduced the ElevenAI app and Scribe v3 Medical, and said it will help businesses fine-tune open-source language models and ElevenLabs voice models in the coming months.
Databricks updates Genie agent to run multi-step business tasks unsupervised, built on a new ontology
Databricks unveiled an updated version of its AI agent Genie, which can now perform multi-step jobs that take extended work to finish. The company says leaders can assign a wide range of tasks and the agent completes them unsupervised. CEO Ali Ghodsi showed TechCrunch how he uses it internally.
Comments
Loading...