Unverified50% confidenceFactExact time
When the model loads, it first allocates a full-precision shell of approximately 1.29 GB and then reads in 643 MB of full weights, with peak memory potentially reaching approximately 3,400 MB.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/25/2026
First Seen
Valid until: 12/24/2026
Sources
Related Entities
Related Claims
VerifiedUsing float16 precision as a baseline, each parameter occupies approximately 2 bytes, meaning a 7B model requires approximately 14GB of VRAM/memory, and a 70B model requires approximately 140GB.77% similarUnverifiedStoring 100 million 1536-dimensional OpenAI embedding vectors in float32 requires approximately 600 GB of memory, which can be compressed to approximately 150 GB after 8-bit scalar quantization.77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/950990API
curl https://kongchang.com/api/v1/knowledge/claims/950990MCP
get_claim(id=950990)