Unverified50% confidenceFactExact time
On newer-generation GPUs that support FP8 compute, such as Hopper and Blackwell, FP8 can achieve nearly double the throughput improvement while using about half the memory of BF16.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedRunning 4-bit weights on Hopper or earlier GPUs usually requires dequantizing back to FP8/FP16 before computation, so the compute advantage of FP4 cannot be obtained.77% similarUnverifiedFP8 is a low-precision computing format natively supported by the Blackwell architecture, including two variants—E4M3 and E5M2—and the Tensor Core's FP8 compute throughput is twice that of FP16.72% similarUnverified现代GPU如NVIDIA H100、H800原生支持FP8计算,理论上可实现约2倍吞吐量提升72% similarUnverifiedConverting from FP32 to FP16 can achieve a 2–4x throughput improvement on the Tensor Cores of modern GPUs.70% similarUnverifiedNVIDIA natively supported the FP8 data type starting from the Hopper architecture, and the Blackwell architecture further added support for FP4.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1051695API
curl https://kongchang.com/api/v1/knowledge/claims/1051695MCP
get_claim(id=1051695)