Unverified60% confidenceFactExact time
NVIDIA's Blackwell architecture (B100/B200) natively supports matrix operations in MXFP4 format at the hardware level, with throughput nearly doubling compared to FP8.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
8/7/2026
First Seen
Valid until: 11/5/2026
Sources
Q8_K_XL与MXFP4命名之争：FP8≠Q8_0的量化真相
Related Claims
VerifiedNVIDIA designed a dedicated Tensor Core instruction mma.kind::mxf4 for NVFP4 in the Blackwell architecture (SM 100/103, such as the B100 and B200 series), and adopted an FP8 scaling factor mechanism with one scale per group of 16 elements.79% similarVerifiedThe NVIDIA Blackwell architecture (GB200/B200 series) introduces native FP4 tensor cores at the hardware level79% similarUnverifiedNVIDIA在Ampere架构（RTX 30系，2020年）引入专用INT8 Tensor Core使INT8推理速度相比FP16提升约2倍，NVFP4是为Blackwell（RTX 50系，2025年）定制的4位浮点格式77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/699425API
curl https://kongchang.com/api/v1/knowledge/claims/699425MCP
get_claim(id=699425)