NVIDIA QM8700 (Quantum HDR)
40 × 200G in 1U. That is 8 Tb/s of switching through 40 QSFP56 cages.
- Logical ports40 × 200G
- Physical cages40 × QSFP56
- Switching capacity8 Tb/s
- Rack units1U
- Typical power300 W
- Maximum power370 W
- Fabricinfiniband
- ASICNVIDIA Quantum
Breakout
A port faster than the endpoint it serves can be broken out when the speeds divide cleanly. That raises the usable port count. Two-tier capacity goes as the square of the port count, so it moves the tier boundary a long way.
| Endpoint speed | Breakout | Usable ports |
|---|---|---|
| 200G | none | 40 |
| 100G | 2:1 | 80 |
| 25G | 8:1 | 320 |
Roles and optics
- Suitable for: compute, storage fabrics
- Cage type: QSFP56. A module of any other form factor will not seat
- Accepted optic coding: nvidia, universal
- Management: Internally managed
- Part number:
MQM8700-HS2F
Coding matters more than it looks. A switch may refuse a transceiver that is not coded for it, and that is one of six axes that have to agree before a link comes up. The others are form factor, reach, fibre type, fibre connector and ferrule polish. See optics compatibility.
Notes
Previous-generation HDR. One port per cage, so optic counts equal port counts here. A 40-port radix caps a 2-tier fabric at 800 endpoints. Verify against the current vendor datasheet before quoting.
Source: NVIDIA Quantum HDR QM8700 documentation, checked 2026-08. Ports, speed, and rack units confirmed; power figures could not be sourced and remain estimates. Needs verification 2026-08-06