#multimodal-model

[ follow ]
#open-source #code-assistant #video-understanding #audio-separation #prompt-based-editing
Artificial intelligence
fromTechCrunch
1 week ago

China's Moonshot releases a new open-source model Kimi K2.5 and a coding agent | TechCrunch

Kimi K2.5 is an open-source multimodal AI trained on 15 trillion visual and text tokens that excels in coding, video understanding, and agent orchestration.
Artificial intelligence
fromTheregister
1 month ago

Meta introduces new SAM AI able to isolate and edit audio

SAM Audio separates and edits audio using text, time, and visual prompts to isolate or remove specific sounds with intuitive, prompt-driven controls.
[ Load more ]