Beyond Multimodal: Building Full‑Modal AGI Via Open‑Weight
Talkat KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026 · 2026-09-07..09-09·
summary
Describes building a full-modal AGI by combining open-weight models across modalities (vision, audio, speech, text, robotic control) and using an open-weight approach to enable composability and rapid iteration. Covers practical techniques for connecting models (e.g., vision and speech models) for end-to-end pipelines, methods for modular training and fine-tuning, and how open-weight ecosystems accelerate research and deployment in AGI development.
AI-written summary · about
related talks
15mMulti-Modal AI Optimization: Combining Text, Vision, and Speech Efficiently
20mMultimodal language models
27mThe Cutting Edge of AI: From LLMs to Multimodal Intelligence
24mMulti-Modal AI Agents Explained | Open-Vocabulary Alignment
40mKeynote: Open-source Multimodal AI in the Wild
40mThe Marvelous Magic of Multimodal AI
48mNavigating the Divide: Open vs. Proprietary AI
01h 11mThe Move Toward AGI: Why Large Language Models Surprised Almost Everyone... | Theoretically Speaking