NVIDIA's H200 Tensor Core GPU, announced in November 2023, integrates 141 GB of HBM3e high-bandwidth memory on package, the highest capacity of any NVIDIA datacenter GPU announced at the time, aimed at serving and training the largest generative-AI models, where GPU memory capacity is frequently the binding constraint on model size and batch size. GPU memory-capacity records in the AI-accelerator market have been advancing roughly annually as vendors compete on this specific metric.
The question, scope, and sources behind this Registry record.
NVIDIA's H200 Tensor Core GPU, announced in November 2023, integrates 141 GB of HBM3e high-bandwidth memory on package, the highest capacity of any NVIDIA datacenter GPU announced at the time, aimed at serving and training the largest generative-AI models, where GPU memory capacity is frequently the binding constraint on model size and batch size. GPU memory-capacity records in the AI-accelerator market have been advancing roughly annually as vendors compete on this specific metric.
The on-package high-bandwidth memory (HBM3e) capacity of NVIDIA's H200 Tensor Core GPU, the highest memory capacity of any commercially announced NVIDIA datacenter GPU at its November 2023 announcement.
Change a parameter to stress-test whether a proposed result is still inside the published specification. This is an audit aid, not a proof checker.
This record has no editable parameters. Read the formal question and assumptions before challenging it.
Current frontiers derived from accepted Claims.
An observed or demonstrated result; no opposing bound is implied.
The frontier is not sacred
Most progress starts with a disagreement that survives contact with evidence. If you can push the known lower bound up or pull the upper bound down, show us the work.
≥ when you have shown that at least this value is achievable.≤ when you have shown that anything above this value is impossible.No vibes. State the value, define the scope, and link the paper, proof, code, or reproduction that lets another person check it. Editors review every challenge before the public record changes.
Challenge this recordAssertions tied to evidence, attribution, and review.
The frontier as it changed over time.
Only accepted Claims matching the current specification contribute to the displayed bounds. Strict inequalities remain open; contradictory Claims require editorial review.
1 accepted Claim, with 1 linked evidence records.
Permanent ID limitsregistry.com/limits/LR-NVIDIA-H200-HBM-CAPACITY
No active verified bounties are linked to this Limit.
View verified bounty tracker ↗No accepted machine-checked reproductions are recorded for this Limit.
Limits Registry. LR-NVIDIA-H200-HBM-CAPACITY. Highest on-chip memory capacity in a commercial datacenter GPU. 2026.