Passage EVK100
Exascale Connectivity for AI
Passage EVK 100 represents a giant leap in photonic interconnect performance. Designed for the extreme demands of scale-up and scale-out AI networks, it delivers massive bandwidth density and unprecedented energy efficiency.
Aggregate I/O
3.2
Tbps aggregate bandwidth @ 112G PAM4
Efficiency
3.4
pJ/bit target link efficiency @ 112G PAM4
Innovation
The CPO Scaling Solution
Passage EVK100 is engineered to meet the bandwidth explosion of exascale AI superpods. By integrating advanced Photonic Integrated Circuits (PIC) with high-performance optical SerDes chiplets in a 2D multi-chip package, Passage EVK 100 de-risks the most aggressive roadmaps in the industry.
Architecture
Optimized for Bandwidth Density
Passage EVK100 uses a 16-wavelength 112G PAM4 DWDM architecture to maximize bandwidth per fiber. The system supports 1.6 Tbps unidirectional (separate Tx/Rx fibers) transmission on a single-mode optical fiber.
Designed for scale-up and scale-out AI networks, the platform leverages high baud rates to maximize link bandwidth density.
Link Architecture
1,600 Gbps per fiber enabled by a 16-channel DWDM O-band link architecture.
Passage EVK100
Ready-to-Deploy Photonics
The Passage EVK100 reference platform demonstrates rack-scale photonic integration, with our state-of-the-art photonic links ready for integration into CPO solutions today.
DWDM Fiber Interface
Sixteen-wavelength optical link reference platform.
Server-Level CPO Evaluation Platform
Dual-board evaluation platform for validating Passage photonic link performance at scale.
Rack-Level CPO Evaluation Platform
Open-standard servers and racks operating at rack-scale. Sampling now.
Technical Specifications
The Future of Exascale
Passage EVK100 is the standard-bearer for AI infrastructure. Contact our sales team for a deep-dive on the technology and integration roadmaps.