Expired75% confidenceFactExact time
FP8 is an 8-bit floating-point format natively supported by NVIDIA's Hopper architecture (H100), which preserves better dynamic range compared to INT8 quantization.
3
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
vLLM Deep Dive: How PagedAttention Enables High-Throughput LLM Inference
githubvllm-project6/6/2026
Related Claims
UnverifiedFP8量化将模型权重从FP16压缩为8位浮点表示，在NVIDIA Hopper架构（H100/H200）上可获得原生加速，且相比INT8对精度损失更小77% similarUnverifiedFP8是NVIDIA Blackwell架构原生支持的数据类型，相比INT量化保留了浮点数的动态范围表达能力77% similarVerifiedFP8 was first introduced with hardware support by NVIDIA in the H100 architecture, with two subtypes: E4M3 and E5M2.75% similarUnverifiedFP8由NVIDIA在Hopper架构（H100）中引入硬件原生支持，随后在Ada Lovelace（RTX 40系）和Blackwell（RTX 50系）架构中持续完善74% similarUnverifiedNVIDIA natively supported the FP8 data type starting from the Hopper architecture, and the Blackwell architecture further added support for FP4.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/54048API
curl https://kongchang.com/api/v1/knowledge/claims/54048MCP
get_claim(id=54048)