The NVIDIA/Mellanox MMS4A00 (980-9IAH1-00XM00) Compatible OSFP Optical Transceiver is a twin-port OSFP IHS/Closed Finned Top module delivering 2x 800Gb/s (1.6T) for InfiniBand XDR networks. It features a single-mode, silicon photonics–based architecture with eight parallel channels, supporting 2×DR4/DR8 transmission. Each 800G port terminates with a 4-channel MPO-12/APC connector. The module uses 200G-PAM4 electrical modulation and provides reliable short-reach connectivity up to 500 metres over eight single-mode fibres. Built with a Broadcom 3nm DSP, this transceiver ensures ultra-low bit error rates and stable, high-speed data delivery in high-density deployments. Its IHS/Closed Finned Top design enables effective thermal management for Quantum-X800 switches, supporting both air- and liquid-cooled configurations. Seamlessly pairing with NVIDIA B300 and DGX systems, the module delivers scalable, low-latency GPU-to-GPU and GPU-to-storage connectivity, ideal for AI, HPC, and large-scale data centre interconnects, unlocking the full potential of high-performance computing fabrics.. Note: "Finned-top" transceivers are now called "Integrated HeatSink" (IHS) as the cooling fins are integrated into the OSFP connector metal top. Commonly used in air-cooled switches. "Flat-top" transceivers are now called "Riding Heat Sink" (RHS) as the cooling fins ride on top of the network card or systems internal cages. This is used for both air-cooled and liquid-cooled applications for DGX, ConnectX-7, and BlueField-3 DPUS.
Highlights
01Built-in Broadcom 3nm DSP Chip, Max. Power Consumption 25W
02SiPh-Based Technology for Lower Power, Cost and Higher Density
03100% Verified, Compatible with NVIDIA Quantum-X800 Switches
04Designed for DGX GB300/B300 Architectures
058x 200G-PAM4 Electrical to Dual 4x 200G-PAM4 Optical Parallel
06Interoperates with OSFP224-DR4-800G-FL for 2x 800G NIC
07Compliant with Hot Pluggable OSFP MSA, IEEE 802.3dj and RoHS
08Class 1 Laser Safety Compliant and DDM Supported
Specifications
26 fields
NVIDIA/Mellanox InfiniBand<br>Compatible
MMS4A00
Vendor Name
FS
Form Factor
Twin-port OSFP IHS/Closed Finned Top
Max. Data Rate
1600Gbps (8x 200Gbps)
Wavelength
1310nm
Max. Cable Distance
500m
Connector
Dual MTP/MPO-12 APC
Media
SMF
Transmitter Type
CW (SiPh Based)
Receiver Type
PIN
TX Power
-3.3~4dBm
Minimum Receiver Power
-6.3dBm
Power Budget
3dB
Receiver Overload
4dBm
Max. Power Consumption
25W
Extinction Ratio
3.5dB
DDM/DOM
Supported
Commercial Temperature Range
0 to 70°C (32 to 158°F)
Modulation (Electrical)
8x 200G-PAM4
Modulation (Optical)
Dual 4x 200G-PAM4
CDR (Clock and Data Recovery)
TX & RX Built-in DSP
Inbuilt FEC
Yes
Packaging Technology
COB (Chip on Board) Packaging
Bit Error Ratio (BER)
6E-12 (FEC Enabled)
Protocols
OSFP MSA, CMIS 5.1, IEEE P802.3dj, OIF 224G-VSR
Warranty
5 Years
Features
Boosting B300 AI Cluster Performance with 1.6T OSFP224 Modules
In the NVIDIA B300 architecture, the 1.6T OSFP optical module provides high-bandwidth, low-latency connectivity between Quantum-X800 InfiniBand switches and DGX B300 servers, ensuring stable, high-speed interconnections, accelerating large-scale distributed AI training, and enhancing overall cluster efficiency. Learn more about the <a href="/de-en/solutions/1.6t-optical-interconnect-for-infiniband-b300-ai-clusters-s0101.html" >B300 InfiniBand solution</a>.
100% Tested by NVIDIA Environment
The FS 1.6T OSFP DR8 modules are tested on NVIDIA Quantum-X800 original devices to ensure the perfect blend of product performance and reliability. Including hardware compatibility (plugging, unplugging, rebooting), software compatibility (connectivity, parametres). Check compatibility with our <a href="/de-en/tool/compatibility-tool-h0026">Optics-to-Device Compatibility Tool</a>.<br/>
Broadcom 3nm DSP & SiPh-Based Technology for Superior Energy Efficiency
Powered by Broadcom's 3nm DSP and silicon photonics chip, this 1.6T XDR transceiver delivers low power consumption of 25W (over 25% power savings) and a tight 6E-12 BER (with FEC Enabled). It enables higher port density, reduced cooling requirements, and lower total cost of ownership.
XDR vs. NDR: Ultra-Large AI Cluster Efficiency Breakthrough
FS's XDR solution featuring the 1.6T transceiver boosts cluster energy efficiency through high bandwidth density and an optimised link architecture. In 512-node configurations, this solution enables unmatched throughput, advanced scalability, significantly reduced power and space requirements and a 40.52% cut in total costs.
Comprehensive Testing Increases Reliability
Qualified through a rigorous process with advanced equipments to ensure you get the high quality and reliable optics.
