Model comparison
Moonshine Base vs Claude 4 Sonnet
Capability radar, value bars, and side-by-side posture.
Moonshine Base
Not scored
value score
Claude 4 Sonnet
91
value score
Moonshine Base ctx
—
Claude 4 Sonnet ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- Claude 4 Sonnet91
Attribute tape
| Field | Moonshine Base | Claude 4 Sonnet |
|---|---|---|
| Developer | Vokra | Anthropic |
| Context | — | 200K |
| Modalities | Text, Code, Image | |
| Openness | — | Proprietary Model |
| Speed | — | Fast |
| Price | — | $3/1M in · $15/1M out |
| Value score | — | 91 |
| Summary | Moonshine Base is a lightweight English automatic speech recognition (ASR) model from Useful Sensors / Vokra, optimized for real-time, low-latency transcription on resource-constrained and edge devices. It uses an encoder-decoder Transformer with RoPE and is reported to outperform similarly sized Whisper models on word error rate while requiring less compute. | Balanced Claude model for coding and agents. |