Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
Scroll to compare models →
| 3 models | fal-aiDiffRhythm: Lyrics to SongRemove × | abr-aiabr-ai/niagara-84m-batch.enRemove × | mistralaiVoxtral-Mini-3B-2507Remove × |
|---|---|---|---|
| What’s different | DiffRhythm is a blazing fast model for transforming lyrics into full songs. It boasts the capability to generate full songs in less than 30 seconds. | 0.084B parameters · 1 language · Encoder: State-space model · Decoder: CTC | Voxtral Mini is an enhancement of Ministral 3B, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and… |
| Date | Mar 4, 2025Listed | Oct 17, 2026Discovered | Jul 19, 2025Listed |
API prices USD · provider configuration shown below | |||
| Listed rate | See provider quotes | See provider quotes | $16.6667 in / $0.00 out/ 1M tokensOpenRouter · all routes ↗ |
| Provider endpoints | 0 endpoints → | 0 endpoints → | 2 endpoints → |
Benchmarks Each result retains its evaluation configuration | |||
| AMI-Cleaned WER ↗English short-form; normalization unversioned | Not reported | 12.01% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 13.57% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| Earnings22-Cleaned-AA-chunked WER ↗English short-form; normalization unversioned | Not reported | 10.83% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 7.66% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| Gigaspeech-Cleaned WER ↗English short-form; normalization unversioned | Not reported | 10.41% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 8.7% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| LS Clean WER ↗English short-form; normalization unversioned | Not reported | 3.09% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 1.48% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| LS Other WER ↗English short-form; normalization unversioned | Not reported | 7.6% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 3.63% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| SPGISpeech WER ↗English short-form; normalization unversioned | Not reported | 2.61% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 2.15% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| URGENT2024 WER ↗English short-form; normalization unversioned | Not reported | 20.16% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 15.66% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| Voice Arena Monsoon WER ↗English short-form; normalization unversioned | Not reported | 8.82% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 4.9% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
| Voxpopuli-AA-Cleaned WER ↗English short-form; normalization unversioned | Not reported | 4.33% WER ↗abr-ai/niagara-84m-batch.enLeaderboard entry | 2.22% WER ↗mistralai/Voxtral-Mini-3B-2507Leaderboard entry |
Specifications | |||
| Context window | — | — | 32.8K |
| Max output | — | — | — |
| Parameters | — | 0.084B | 3B |
| Audio input | Not reported | Supported | Supported |
| Transcription | Not reported | Supported | Supported |
| Open weights | Not reported | Not reported | Supported |