Cloud GPUs are sold out, as of NVIDIA's third quarter fiscal 2026 results (Nov. 2025).
Mixed Partly true and partly false in roughly equal measure.
Factual: Statements about the present or the past, checked against the record: did they hold up?
The second judge said resolved false; the verdict shown is the first judge's.
Said for an organisation they lead or work for; it counts on both records.
What would count as true or false
Judged as of November 2025 (the Q3 FY2026 results, 2025-11-19). True: contemporaneous evidence shows cloud GPU capacity, especially NVIDIA data-center GPUs at the major cloud providers, was fully booked or unavailable to new demand at that time (e.g., cloud providers or analysts reporting capacity sold out or supply-constrained, with no meaningful spare inventory). Mostly true: broadly sold out or severely constrained at major providers, with some spare capacity at smaller providers or for older GPU generations. Mixed: some leading clouds or GPU types were sold out while others had readily available capacity. Mostly false: capacity was tight only in pockets, and on-demand cloud GPUs were generally available. False: evidence shows widespread available capacity, falling rental prices driven by oversupply, or GPUs readily rentable on demand across major clouds.
Set before the research began.
Evidence (4)
-
E1 Nvidia chips sold out? Cut back on AI plans, or look elsewhere Network World · 20 November 2025
“Nvidia CFO Colette Kress’s claim that “The clouds are sold out, and our GPU-installed base […] is fully utilized,””
-
E2 Nvidia chips sold out? Cut back on AI plans, or look elsewhere Network World · 20 November 2025
“Among the doubters is Forrester senior analyst Alvin Nguyen, who said Thursday he has had clients ask what to do when addressing the lack of available Nvidia GPUs, since demand far exceeds supply.”
-
E3 Clouded Judgement 10.31.25 - Cloud Giants Report Q3 Clouded Judgement (Substack newsletter) · 31 October 2025
“Most importantly - they ALL called out still being meaningfully capacity constrained.”
-
E4 Is this a data center bubble, not an AI bubble? - FXStreet FXStreet · 19 November 2025
“SemiAnalysis, which accurately predicted this decline in December 2023, now characterizes it as "a buyers' market for GPU rentals" with "widespread availability from over 100+ AI Neoclouds and Hyperscalers."”
Reasoning
An independent newsletter dated 2025-10-31 reports that all the major cloud providers still described themselves as meaningfully capacity constrained (E3). A Forrester analyst says clients face a lack of available Nvidia GPUs because demand far exceeds supply (E2). However, a 2025-11-19 FXStreet piece says SemiAnalysis "now characterizes" the market as "a buyers' market for GPU rentals" with "widespread availability from over 100+ AI Neoclouds and Hyperscalers" (E4), which points against a full sell-out, though the piece relays a characterisation rather than measuring capacity. E1 only relays Nvidia's own CFO statement, so it adds no independent support. The evidence points both ways, so the claim resolves as mixed.
Cite this claim
Jensen Huang, “Cloud GPUs are sold out, as of NVIDIA's third quarter fiscal 2026 results (Nov. 2025).” (said November 2025). Verdict: mixed. Reliability API, claim c_712affa249a0, method 0.1.0+03d23c0e, dataset release v1.1.2. https://epistacklabs.com/claims/c_712affa249a0
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
More factual claims by Jensen Huang
-
Mostly true
“we invented a technology called a GPU, and basically a GPU does parallel processing tasks”
-
Resolved false
“Today, NVIDIA’s token cost is the lowest in the world, not by 10%, by X factors, orders of magnitude, all because we did extreme co-design”
-
Mostly true
“It’s the first one to be PCI Express Gen 6.”
-
Mixed
“We have demonstrated that NVFP4 can achieve inference without loss of precision, while delivering significant performance and energy efficiency improvements, and it is equally suitable for training.”
-
Mixed
“Over the past two years, the computational demands for inference have increased approximately 10,000 times, while usage has increased approximately 100 times.”
Verdicts are made automatically; how verdicts are made.