Unverified50% confidenceFactExact time
FP8 is a format that stores weights and/or activations in 8-bit floating point numbers, with VRAM and memory usage approximately twice that of 4-bit quantization.
1
Sources
50%
Confidence
Long-term
Relevance
9/25/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFP8相比BF16可提供约2倍的内存带宽节省和更高的矩阵计算吞吐78% similarPartially VerifiedA standard FP32 model occupies 4 bytes per parameter, while INT4 quantization compresses this to 0.5 bytes, theoretically achieving an 8x memory saving.74% similarVerifiedQuantized KV caches (such as storing KV vectors in FP8 or INT4) can compress the cache size to 1/2 or even 1/4 of the original.72% similarUnverifiedFP4 相对 FP16 可节省75%存储，采用每组16或32个权重共享一个 FP8 缩放因子的分组量化策略71% similarVerifiedINT4 achieves 4× storage compression compared to FP16.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/949599API
curl https://kongchang.com/api/v1/knowledge/claims/949599MCP
get_claim(id=949599)