Unverified50% confidencePredictionExact time
The inference throughput of the B200 (Blackwell architecture) is expected to be several times that of the H100.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/7/2026
First Seen
Valid until: 12/6/2026
Sources
Related Entities
Related Claims
UnverifiedCompared to the previous-generation H100, the B200 achieves approximately a 5x generational improvement in inference performance.75% similarUnverifiedH100内存子系统的动态功耗在大模型推理中占比可达整卡功耗的20%～35%71% similarUnverified与H100相比，Blackwell在大语言模型推理性能上提升约30倍，能效提升约25倍65% similarUnverified英伟达H100相比A100算力提升约3倍，能效比提升约5倍65% similarUnverifiedMainstream GPU inference solutions based on the NVIDIA H100 can typically only achieve tens to one or two hundred tokens per second on models of equivalent scale.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/870924API
curl https://kongchang.com/api/v1/knowledge/claims/870924MCP
get_claim(id=870924)