| What it does | Small model. Big decisions.
| Voice assistant for AI coding agents on Mac, Windows and Linux
|
|---|
| Stage | Not given
| Startup
|
|---|
| Founded | Not given
| Not given
|
|---|
| Based in | CA
| IN
|
|---|
| Pricing model | Freemium
| Freemium
|
|---|
| Pricing | Test plan: 125 million free input tokens per month with test_sk keys. Production plan: $0.04 per million input tokens with prod_sk keys (no charge for output tokens). Minimum credit bundle of $10 to unlock production access.
| Free forever for voice coding and meeting transcription. AgentCall meeting feature: per-minute pricing with free hours included.
|
|---|
| Key features | - Classification with score distribution and confidence
- Text and image extraction
- Yes/no decisions with probability
- Entity recognition and extraction
- Field verification against source
- Multi-level classification (classify-tree)
- OpenAI SDK compatibility
- Sentiment or tone rating
- Question answering with source spans
- SDK support for TypeScript and Python
| - Local speech-to-text and neural voice synthesis on-device
- Full-duplex echo cancellation for uninterrupted conversation
- Approve-before-send review mode for transcripts
- Multi-project support with voice per project
- Screenshot capture and sharing with agent requests
- Google Calendar integration for automatic meeting joining
- Local meeting recording without bot participant
- Support for 6+ coding agents including Claude Code and Cursor
- Global hotkeys and mute controls
- Supports 6+ languages including English, Spanish, French, Italian, Portuguese and Hindi
|
|---|
| What makes it different | - Optimized for bounded decisions with millisecond latency, avoiding slow token-by-token generation
- Extremely cost-effective at $0.04 per million input tokens with no output token charges
- 125 million free test tokens monthly with no card required
- Production-ready with REST API, SDKs, and CLI interfaces
- Structured responses enable direct application logic without additional parsing
| - All processing happens on your machine with zero cloud uploads
- Two-way voice conversation loop, not just dictation
- Works completely offline with no internet connection required
- Free forever for core voice and transcription features
- Live agent status indicator showing real-time listening state
|
|---|
| Free plan | Not given
| Not given
|
|---|
| Open source | Not given
| Not given
|
|---|
| Platforms | Not given
| Not given
|
|---|
| Public API | Not given
| Not given
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 18 ● no change since last check | 18 ● no change since last check |
|---|
| Trust Flow (Majestic) | 28 | 8 |
|---|
| Citation Flow (Majestic) | 31 | 26 |
|---|
| Referring domains (Majestic) | 42 | 171 |
|---|
milliseconds.ai
SKI