

Kinetix
Reward (1.00 QUA)3D-Conditioned AI Video Model
At Kinetix, we’re building the next generation of AI video models — unlocking full creative control through 3D-conditioned generation.
Our technology enables creators to generate videos where both human motion and camera movement are powered by 3D under the hood, combining cinematic quality with precise, editable control.
This breakthrough builds on 5 years of advanced R&D in 3D foundation models, giving us a unique edge to bring character-driven storytelling into the era of generative video.
Watch our video review: Kinetix 2026: 3D-Conditioned AI Video with Real Motion Intelligence
Just released


TwelveLabs
Building the world's most powerful video understanding platform.
Reward (1.00 QUA)

Hippocratic AI
Hippocratic AI - A safety focused large language model (LLM) for the healthcare industry
Reward (1.00 QUA)

Suki
Building presence at the heart of care.
Reward (1.00 QUA)

Abridge
#1 Best in KLAS enterprise-grade AI for healthcare. Patient-centered clinician intelligence for every moment of care.
Reward (1.00 QUA)