Unverified50% confidenceFactExact time
The decommissioned supercomputer is being sold as a package in the form of 300 racks, with a GPU-to-CPU ratio of roughly 2:1, a typical high-density AI training node configuration.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/28/2026
First Seen
Valid until: 11/11/2026
Sources
Related Entities
Related Claims
UnverifiedOn an A100 (80 GB VRAM) GPU, the throughput for training a 7B model is approximately 3,000–5,000 tokens per second.71% similarVerifiedRunning local AI models typically requires 8–16 GB or more of memory, a relatively recent CPU or a GPU-accelerated graphics card, and sufficient storage space.71% similarUnverifiedAfter all 31 layers of the 26B model are offloaded to the GPU, ollama ps still reports 14%/86% CPU/GPU, which originates from mmap-mapped CPU-side address space tensor buffers rather than actual CPU computation.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1108906API
curl https://kongchang.com/api/v1/knowledge/claims/1108906MCP
get_claim(id=1108906)