Gemini 3.1 Flash Live: Google's latest audio model with improved benchmarks and watermarking
What's new in Gemini 3.1 Flash Live
Google has released Gemini 3.1 Flash Live, their highest-quality audio and voice model designed for real-time dialogue. The model delivers improved speed and natural rhythm for voice-first AI applications.
Key technical details
- Benchmark scores: 90.8% on ComplexFuncBench Audio (multi-step function calling with constraints) and 36.1% on Scale AI's Audio MultiChallenge (complex instruction following with "thinking" on)
- Improved capabilities: Better tonal understanding, recognition of acoustic nuances like pitch and pace, and dynamic adjustment to user frustration or confusion
- Watermarking: All audio generated includes SynthID watermark for AI content detection
- Multilingual support: Available in over 200 countries and territories
Availability and access
- For developers: Available in preview via Gemini Live API in Google AI Studio
- For enterprises: Included in Gemini Enterprise for Customer Experience
- For general users: Accessible via Search Live and Gemini Live
The model enables building voice-ready agents that handle complex tasks in noisy environments and supports longer conversation threads during extended interactions.
📖 Read the full source: HN AI Agents
👀 See Also
Anthropic Doubles Claude Code Rate Limits, Removes Peak Throttling for Paid Plans
Anthropic has doubled 5-hour rate limits for Claude Code across Pro, Max, Team, and Enterprise plans, removed peak-hour throttling, and boosted API rate limits for Opus models.
Claude Code Deletes Production Database After Terraform State File Error
A developer used Claude Code to manage AWS infrastructure with Terraform, but a missing state file led to duplicate resources and a subsequent 'destroy' operation that wiped 2.5 years of records including database snapshots.
Claude Opus 4.6 effort=low parameter causes lazy agent behavior
When using effort=low with Claude Opus 4.6, agents made fewer tool calls, were less thorough in cross-referencing, and ignored parts of system prompts about web research. Switching to effort=medium resolved the issues.
Claude's speech recognition limitations and user workaround with Spokenly and Parakeet TDT
A user reports Claude's built-in microphone transcription is inaccurate compared to ChatGPT's, creating more work than it saves. They implemented a workaround using Spokenly on Mac with NVIDIA's Parakeet TDT model for improved performance.