Claude Code adds voice mode for hands-free coding commands
Anthropic has launched voice mode for Claude Code, its AI coding assistant for developers. The feature enables hands-free, conversational coding workflows through spoken commands.
Key details from the source
According to engineer Thariq Shihipar's announcement on X:
- Voice mode is currently live for approximately 5% of users
- Broader rollout is planned for the coming weeks
- To enable: type
/voiceto toggle it on, then speak your command - Example command: "refactor the authentication middleware"
- Users will see a note on the welcome screen once they have access
The source notes several unknowns about the implementation:
- Limitations of the capability are unclear
- It's unknown if there are caps on voice interactions or specific technical constraints
- It's unclear if the feature was built with a third-party AI voice provider like ElevenLabs
Anthropic previously launched voice mode for its standard Claude chatbot in May, allowing voice interaction for general-purpose tasks. The company reported in February that Claude Code's run-rate revenue surpassed $2.5 billion, more than doubling since the beginning of 2026, with weekly active users doubling since January.
📖 Read the full source: HN AI Agents
👀 See Also
Benchmark Comparison of Qwen 3.5 Models Against Major AI Models
A benchmark comparison website includes verified scores and head-to-head infographics for Qwen 3.5 models (122B, 35B, 27B, 397B) against models like GPT-5.2, Claude 4.5 Opus, Gemini-3 Pro, and others.
Ångstrom Used Claude Code to Train a Model That Beat Meta's UMA-OMC — 100k GPU Jobs on Spot
Ångstrom (YC S24) trained CSP-MACE-Å, an ML model 10,000x faster than DFT with matching accuracy, outperforming Meta's UMA-OMC on crystal structure prediction. They used Claude Code to orchestrate 100,000 GPU jobs on multi-cloud spot via Anycloud CLI.
OpenClaw Experiment: AI Agents Choosing Silence to Improve Signal-to-Noise Ratio
An OpenClaw experiment gives AI agents autonomy to skip tasks when they can't add value, logging silence decisions to a 'silence log' with reasoning. The system uses LLM calls before content generation and auto-adjusts thresholds after 3 consecutive silence days.
Meta Releases BOxCrete AI Model for Concrete Mix Design
Meta has released Bayesian Optimization for Concrete (BOxCrete), an open-source AI model for designing sustainable concrete mixes using U.S.-produced materials. The model improves on previous versions with better noise robustness and slump prediction capabilities.