Unverified60% confidenceFactExact time
GPU HBM bandwidth is typically on the order of 2–3 TB/s (3.35 TB/s for H100), while the effective bandwidth of Host DRAM connected via PCIe 5.0 is only about 64 GB/s—a difference of nearly 50x.
2
Sources
60%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
TB/s内存层加速SGLang：X-Mem让TTFT降低6.7倍的技术解析
twitterlmsysorg7/8/2026
Related Claims
VerifiedA100 GPU的HBM带宽约2TB/s，而片上SRAM带宽可达19TB/s，相差近10倍83% similarUnverifiedSwapping KV cache in and out depends on the interconnect bandwidth between host memory and GPU VRAM; taking PCIe 4.0 x16 as an example, the one-way effective bandwidth is on the order of about 25 GB/s, while the HBM bandwidth of a single data center GPU can reach 2–3 TB/s, a difference of about two orders of magnitude.79% similarUnverifiedThe HBM3 memory bandwidth of the NVIDIA H100 is as high as 3.35TB/s, while the Ryzen AI Halo, constrained by LPDDR5x bandwidth, typically operates in the range of 100–300GB/s.79% similarUnverifiedA single HBM3E stack can contain 8 to 12 layers of DRAM die, providing over 1 TB/s of bandwidth.79% similarUnverifiedNVIDIA's H100 is equipped with 80 GB HBM3, delivers memory bandwidth of 3.35 TB/s, and NVLink 4.0 provides 900 GB/s bidirectional bandwidth.79% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/486856API
curl https://kongchang.com/api/v1/knowledge/claims/486856MCP
get_claim(id=486856)