MoshiVis is a Vision Speech Model built as a perceptually-augmented version of Moshi v0.1 for conversing about image inputs
Kyutai
non-profit
Verified
AI & ML interests
None defined yet.
Recent Activity
Collections
3
spaces
1
models
30
kyutai/moshi-artifacts
Updated
kyutai/moshika-vis-pytorch-bf16
Updated
•
55
kyutai/moshika-vis-mlx
Updated
•
2
kyutai/moshika-vis-candle-q8
Updated
•
8.31k
kyutai/moshika-vis-candle-bf16
Updated
•
1
kyutai/hibiki-1b-rs-q6k
Updated
•
16
kyutai/hibiki-1b-rs-q8
Updated
•
15
kyutai/hibiki-2b-rs-bf16
Translation
•
Updated
•
3
kyutai/hibiki-1b-rs-bf16
Translation
•
Updated
•
9
kyutai/hibiki-2b-pytorch-bf16
Translation
•
Updated
•
119
•
50