Unverified50% confidenceConceptExact time
NVFP4 is a block-scaled 4-bit floating-point format defined by NVIDIA, where every 16 numbers share one FP8 scale; MXFP8 is an 8-bit block floating point defined by OCP, where every 32 numbers share one E8M0 scale.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
Partially VerifiedNVFP4 occupies 4 bits per parameter, i.e., 0.5 bytes, theoretically halving weight memory usage and load bandwidth compared with FP8.74% similarVerifiedThe MXFP4 format has a group of elements (typically 32) sharing a single 8-bit scaling factor, with each element represented by a 4-bit micro floating-point number (1 sign bit + 2 exponent bits + 1 mantissa bit).72% similarUnverifiedNVFP4量化需要50系列NVIDIA显卡才能使用67% similarUnverifiedNVIDIA released an NVFP4 quantized model of Qwen3 235B on Hugging Face, compressing the model from approximately 470 GB of VRAM in BF16 to approximately 156 GB, enabling it to run on a single-node cluster of 8 H100 GPUs.67% similarUnverifiedNVFP4格式仅能表示约16个不同量级66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1049824API
curl https://kongchang.com/api/v1/knowledge/claims/1049824MCP
get_claim(id=1049824)