Available Models
7
📦
Downloaded
2
✅
Current Model
None
🎯

Available Models

Model Status Size Parameters Speed Description Actions
Tiny Not Downloaded 39 MB 39M ~32x Fastest, least accurate
Base Downloaded 74 MB 74M ~16x Fast, good for English
Small Downloaded 244 MB 244M ~6x Balanced speed/accuracy
Medium Not Downloaded 769 MB 769M ~2x Good accuracy, slower
Large Not Downloaded 1550 MB 1550M ~1x Best accuracy, slowest
Large V2 Not Downloaded 1550 MB 1550M ~1x Latest large model
Large V3 Not Downloaded 1550 MB 1550M ~1x Newest, most accurate

📋 Model Information

Model Sizes:

  • Tiny: Fastest but least accurate (39 MB)
  • Base: Good for English, fast (74 MB)
  • Small: Balanced performance (244 MB)
  • Medium: Good accuracy (769 MB)
  • Large: Best accuracy, slow (1.5 GB)

💡 Usage Tips

Choosing the Right Model:

  • Use Tiny/Base for real-time transcription
  • Use Small for balanced quality
  • Use Medium/Large for highest accuracy
  • Consider your hardware capabilities