Verified85% confidenceFactExact time
The HBM memory bandwidth of the NVIDIA H100 is approximately 3.35 TB/s, and its utilization rate in autoregressive generation tasks is often below 10%.
5
Sources
85%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
GPT-5.6三模型解析：Sol/Terra/Luna定价与监管困局
bilibili63号炼金工坊7/1/2026
Related Claims
UnverifiedThe 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.84% similarVerifiedThe NVIDIA H20's peak BF16 compute is approximately 296 TFLOPS, significantly reduced compared to the H100's 989 TFLOPS, but it retains 141 GB HBM3 memory and 4.0 TB/s memory bandwidth.83% similarUnverifiedNVIDIA H100搭载的HBM3提供约3.35TB/s内存带宽，而DDR6预计单通道最高带宽约256GB/s81% similarVerifiedThe HBM3 memory used by NVIDIA's H100 has a bandwidth of up to 3.35 TB/s.79% similarUnverifiedGPU 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.78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/487811API
curl https://kongchang.com/api/v1/knowledge/claims/487811MCP
get_claim(id=487811)