Skip to main content

Claude Voice Mode Gets a Major Upgrade: Opus and Sonnet Join the Party

Anthropic is rolling out a significant upgrade to Claude's voice mode this week, according to tech media outlet TestingCatalog. The update will let users choose between three different AI models—Opus, Sonnet, and Haiku—instead of being stuck with just Haiku. That means your voice chats with Claude are about to get a whole lot smarter.

Three Models, One Voice

Currently, Claude's voice mode runs exclusively on the Haiku 4.5 model, which is fine for casual chit-chat but not exactly a powerhouse. With the new model selector, you'll be able to pick the model that fits the conversation. Need a quick answer? Stick with Haiku. Want to dive deep into a complex topic? Switch to Opus or Sonnet for more nuanced responses.

Alongside the model selector, there's a setting to control the "thinking" level—essentially how much effort the AI puts into crafting its reply. This gives users even finer control over the interaction, balancing speed and depth as needed.

What About Claude Fable?

Anthropic's flagship model, Claude Fable (codenamed Mythos), won't be available in voice mode just yet. But given the company's track record, it's likely to appear in a future update. For now, Opus and Sonnet are the heavy hitters in the voice lineup.

Why This Matters

Voice assistants have long been limited to simple tasks—setting timers, playing music, or reading the weather. By offering multiple models with adjustable thinking, Anthropic is pushing voice AI toward more meaningful conversations. Imagine debating philosophy with Opus or brainstorming business strategies with Sonnet—all by speaking naturally.

This upgrade also signals a shift in how AI companies think about voice interfaces. Instead of a one-size-fits-all approach, users get to choose the right tool for the job. It's a move that could set a new standard for voice AI.

Key Points

  • Upgrade this week: Claude's voice mode will support Opus, Sonnet, and Haiku models.
  • New model selector: Users can switch between models and adjust thinking levels.
  • Haiku no longer alone: Previously, only Haiku was available for voice.
  • Claude Fable missing: The flagship model isn't included yet but may come later.
  • Deeper conversations: The update enables more complex and nuanced voice interactions.