Verified65% confidenceFactExact time
On traditional x86 platforms, the bandwidth ceiling of the PCIe bus connecting a discrete GPU to the motherboard memory is approximately 64GB/s.
3
Sources
65%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
苹果M7 Ultra芯片曝光：1.5TB统一内存能否重塑本地AI推理格局
Related Claims
VerifiedThe 3080 Ti has a memory bandwidth of 912 GB/s, far exceeding the 16–32 GB/s PCIe bus bandwidth between CPU and GPU.80% similarVerifiedConventional PCIe Gen5 provides approximately 128 GB/s of bidirectional bandwidth.79% similarVerifiedNVLink 4.0 has a bidirectional bandwidth of up to 900 GB/s, while the bandwidth of a traditional PCIe bus is approximately 64 GB/s.78% similarVerifiedThe unified memory bandwidth of the M3 Max chip is as high as 400GB/s, far exceeding the approximately 64GB/s of a traditional PCIe bus77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/507764API
curl https://kongchang.com/api/v1/knowledge/claims/507764MCP
get_claim(id=507764)