)
Run the Models (Not Just the World)
Dec 10, 2025
Next Generation Stage
Next Generation
- How do you build for AI that sees, hears, and speaks? Multimodal models demand more than compute, they demand a full stack rethink. These systems are redefining how we architect pipelines, optimize hardware, and design interfaces across modalities.
- Gain an in-depth understanding of how high-performance systems are adapting for vision, speech, and embodied AI use cases, with practical insights into latency trade-offs, pipeline orchestration, and runtime architecture.
Session Type
Presentation
Content Focus
Technical
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
)
)
)
)
)
){kind=logo}
)
)
){kind=logo}
)
.svg){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
){kind=logo}
)
.jpg/fit-in/500x500/filters:no_upscale()){kind=logo}
)
.jpg/fit-in/500x500/filters:no_upscale())
){kind=logo}
){kind=logo}
)
)
)
)
)
)
){kind=logo}
)