SpaceX is developing the Kessel AI chip rack to reduce reliance on NVIDIA
SpaceX is developing the Kessel AI chip rack, using a passive PCB middle board to replace the NVLink cable box, retaining 72 GPU modules, 18 computing trays, and 9 switching trays, planning to procure chip modules directly from NVIDIA and purchase the remaining components independently to bypass NVIDIA's L6 computing board and enhance supplier selection and procurement autonomy.
Same story1 other outlet also reported this
- ITHome · AI··First
SpaceX explores 'de-Nvidia hóa' for its server rack layers: Kessel custom AI server rack plan retains Nvidia chips but more auxiliary hardware is self-purchased
- 36Kr · AI··+22 h 40 min·Reading now
SpaceX is developing the Kessel AI chip rack to reduce reliance on NVIDIA
Adds:uses passive PCB mid-boards instead of NVLink cable boxes
In order of publication · later reports are added as they appear · Compared by AI from each outlet’s title and summary
Source: 36Kr · AI · 36kr.com