Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
Scroll to compare models →
| 3 models | intfloatmultilingual-e5-large-instructRemove × | kreaKrea 3 EditRemove × | nvidiaNemotron-3-DiarizationRemove × |
|---|---|---|---|
| What’s different | The Multilingual-E5 models, initialized from XLM-RoBERTa, support up to 512 tokens per input — any longer text will be silently truncated. To ensure optimal performance, always prefix inputs with “query:”… | Edits and combines reference images using text instructions, with optional prompt expansion, output-size controls, and reproducible generation through a seed. | Nemotron 3 Diarization is an open-weight streaming Sortformer model, engineered for low-latency speaker diarization across up to 8 speakers, ordered by arrival time. It powers meeting transcription, call… |
| Date | Apr 17, 2025Listed | Oct 14, 2026Listed | Sep 25, 2026Listed |
API prices USD · provider configuration shown below | |||
| Listed rate | See provider quotes | $0.025/ megapixelfal · all routes ↗Output + reference images | $0.00001/ secondDeepInfra · all routes ↗Input audio second |
| Provider endpoints | 0 endpoints → | 1 endpoint → | 1 endpoint → |
Specifications | |||
| Context window | 512 | — | — |
| Max output | — | — | — |
| Parameters | — | — | — |