Twelve Labs — Feature
Twelve Labs provides multimodal AI models that make video searchable and analyzable through text-based querying and understanding. The page showcases two flagship models—Marengo (multimodal embedding) and Pegasus (video language model)—with technical descriptions and architectural diagrams illustrating how video, audio, image, and text inputs feed into processing pipelines to enable video analysis and discovery.
This is a static, crawlable rendering for search & AI engines. The interactive version — filters, live decode console, copy & export — is at https://mozaika.design/inspiration/twelve-labs-perception-and-orchestration.
Measured spec
| Scheme | dark |
| Background | #1E1C1A |
| Text | #6B685C |
| Contrast | 3.04 |
| Columns | 2 |
| Alignment | split |
| Whitespace | airy |
| Density | 0.138 |
| Measured size | 1600×1488 px |
Palette
#F5F5F5#1D1C1B#5864ED
Fonts
Milling Duplex 1mm
Page types
Detail
UX patterns
Dark Mode
UI elements
Card
Context
Other sections on this page
- Hero — See the Unseen Hero
- Feature — Video Infrastructure Product Overview
- Feature — Video library search and workflow management
- Stats / Metrics — Video processing and compliance hero
- Logo Wall — Customer Logo Row
- Gallery / Showcase — Built for Every Workflow
- Feature — Secure by Design
- Testimonial / Social Proof — Customer Stories
- CTA / Sign-up — Ready to See CTA
- Footer — Footer Navigation