HPC infrastructure reallocation plan
Company: Core Scientific
The claim, verbatim
700 megawatts of infrastructure reallocated for HPC hosting, with 500 megawatts for hosted GPU power and 200 megawatts for HPC support infrastructure.
Source (primary)
Core Scientific, Inc. Investor and Analyst Event June 12, 2024 Transcript (IR page (investors.corescientific.com), earnings_call)
View cached copy (2026-09-26)Live source ↗
Quote: “we aim to provide 700 megawatts for HPC, of which 500 megawatts will be for the hosted GPUs, so that's about 200 megawatts that are needed in order to support those facilities”
How we checked this
Checked on September 28, 2026. The cited source supports every part of this claim.
Both the CEO and the CFO describe roughly 700 MW allocated to HPC, split into 500 MW for GPUs and 200 MW of support. The document presents this as a plan ('we aim to provide').
Confirmed in the source:
- Core Scientific plans to provide about 700 megawatts for HPC
- 500 megawatts of that is for hosted GPU power
- About 200 megawatts is for HPC support infrastructure
What we did: Read our cached copy of IR page (investors.corescientific.com) (https://d1io3yog0oux5.cloudfront.net/_fda3a1f2f1890adaa7ac6c4d9d310971/corescientific/db/1066/11128/pdf/June_2024_Investor_Day_Transcript_061224.pdf) in full (94,603 characters, retrieved September 26, 2026) and checked each assertion in the claim against it.
Additional evidence
confirms Core Scientific, Inc. Investor and Analyst Event June 12, 2024 Transcript
Quote: “we aim to provide 700 megawatts for HPC, of which 500 megawatts will be for the hosted GPUs, so that's about 200 megawatts that are needed in order to support those facilities”