Unverified50% confidenceBenchmarkExact time
Blackwell 的第五代 NVLink、HBM3e 显存和专用 FP4 张量核心使其在推理密集型场景下相较 H100 可获得2-4倍吞吐量提升
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
Related Claims
UnverifiedThe Blackwell architecture introduces fifth-generation Tensor Cores with native support for FP4×FP4 matrix multiply-accumulate operations, using FP32 or FP16 precision for the accumulator.73% similarPartially VerifiedNVFP4 is a 4-bit floating-point (FP4) quantization format introduced by NVIDIA, designed specifically for the hardware acceleration capabilities of Blackwell architecture GPUs68% similarUnverifiedBlackwell introduces FP4 for inference and, in the form of the GB200 NVL72 rack, unifies 72 GPUs into a single compute pool with unified addressing.68% similarUnverifiedBlackwell is NVIDIA's next-generation microarchitecture following Hopper and Ada Lovelace, introducing fifth-generation Tensor Cores to improve FP4/FP6 inference efficiency.68% similarUnverifiedCore innovations of the Blackwell architecture include the second-generation Transformer Engine (supporting FP4 precision), the NVLink Switch system (supporting interconnection of 576 GPUs), 20 petaFLOPS of FP4 compute, and 192 GB HBM3e memory (with bandwidth reaching 8 TB/s).67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/336767API
curl https://kongchang.com/api/v1/knowledge/claims/336767MCP
get_claim(id=336767)