Verified70% confidenceBenchmarkExact time
NVIDIA's H100 features a Transformer Engine specifically designed for AI workloads, delivering nearly 4,000 TFLOPS of compute at FP8 precision, approximately 3 times the improvement over the A100.
4
Sources
70%
Confidence
Long-term
Relevance
7/23/2026
First Seen
Sources
Related Claims
UnverifiedNVIDIA H100 的 Transformer Engine 将 FP8 训练和推理的吞吐量相比上一代提升近 4 倍82% similarUnverifiedNVIDIA A100、Google TPU v4等专用AI训练芯片针对Transformer的矩阵运算模式进行了专项优化81% similarUnverifiedNVIDIA launched the Transformer Engine alongside the H100, which maintains historical maximum absolute value statistics of activations for each tensor and dynamically adjusts scaling factors.80% similarUnverifiedThe Hopper architecture of the H100 introduced a dedicated Transformer Engine supporting FP8 precision computation.79% similarUnverifiedH100的Transformer Engine支持FP8精度，张量核心在FP8模式下峰值算力可达3958 TFLOPS，FP16为1979 TFLOPS79% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/599795API
curl https://kongchang.com/api/v1/knowledge/claims/599795MCP
get_claim(id=599795)