| Eleven v3 | Multilingual v2 | Flash v2.5 | Turbo v2.5 | |
|---|---|---|---|---|
| Latency | Higher | ~250 300ms | ~75ms | ~250 300ms |
| Languages | 70+ | 29 | 32 | 32 |
| Character limit | 5,000 | 10,000 | 40,000 | 40,000 |
| Quality | Highest (most expressive) | Very high (most stable) | Good (optimized for speed) | High (balanced) |
| Credit cost | Full rate | Full rate | ~50% discount | ~50% discount |
| Best for | Character voices, audiobooks | Long form, professional content | Real time agents, games | Chatbots, balanced apps |
Model selection tip
Start with eleven_multilingual_v2 (the default) for most use cases. Switch to Flash v2.5 if you need sub 100ms latency for conversational AI. Use v3 when you need the most expressive, dramatic voice delivery and can work within the 5,000 character limit.