SenseNova 5.5
modelUnveiled at WAIC 2024 in Shanghai. 30% performance improvement over SenseNova 5.0. Introduced SenseNova 5o, China's first real-time multimodal model rivaling GPT-4o's streaming interaction across audio, text, image, and video. Also debuted the Vimi video generation model and reduced edge deployment cost to RMB 9.90 per device per year.
Model Details
Variants
| Name | Parameters | Notes |
|---|---|---|
| SenseNova 5.5 | — | — |
| SenseNova 5o | — | Real-time multimodal streaming interaction model |