Unverified50% confidenceFactExact time
The weight volume of this block-FP8 build is approximately 31 GB, roughly half that of BF16 weights.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe BF16 weights of the 27B model are approximately 54GB (27B × 2 bytes).79% similarUnverifiedA 2.4-trillion-parameter model requires approximately 9.6 TB of VRAM in FP32, about 4.8 TB in BF16, about 2.4 TB in INT8, and about 1.2 TB in INT4.72% similarUnverifiedFP8 quantization can compress the weights of a 7B model from approximately 14GB (FP16) to approximately 7GB.72% similarVerifiedA 70B-parameter FP16 model requires approximately 140 GB of video memory, approximately 70 GB after INT8 quantization, and approximately 35–40 GB after INT4 quantization.71% similarUnverifiedDouble quantization further quantizes the FP32 scaling factors shared among every 64 parameters into FP8 format, reducing the quantization overhead per parameter from 0.5 bits to approximately 0.127 bits.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/830172API
curl https://kongchang.com/api/v1/knowledge/claims/830172MCP
get_claim(id=830172)