VR2-0163
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Rubin Oberon Rack: NVL72 not NVL144 nor NVL36 - Line hint:
207
Context Before
Source: Nvidia VR NVL72 BoM and Power Budget Model ↗
Evidence
On a side note, VR NVL72 was initially known as VR NVL144 as Jensen math ↗ from GTC 2025 defined the number of GPU as the number of GPU compute die in system (with 2 compute dies per package and 72 Rubin packages per Oberon rack = 144 compute die). The naming was changed back to VR NVL72 to represent the 72 Rubin GPU packages in the system in late December. This was right before CES 2026 where the naming was officially confirmed as VR NVL72.
Context After
CPX Form Factor
② Atomic Claim
VR NVL72 最初被稱為 VR NVL144，因為 GTC 2025 的命名方式以系統中的 GPU compute die 數量計算；每個 package 有 2 個 GPU compute dies，Oberon rack 有 72 個 Rubin packages，因此共 144 個 compute dies。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"attribute": "COMPUTE_PERFORMANCE",
"context_nodes": [
{
"id": "04_knowledge_base/VR NVL144",
"label": "VR NVL144"
},
{
"id": "04_knowledge_base/GPU",
"label": "GPU"
},
{
"id": "04_knowledge_base/NVIDIA Oberon",
"label": "Oberon"
},
{
"id": "04_knowledge_base/Vera Rubin",
"label": "Rubin"
}
],
"entity": {
"id": "04_knowledge_base/VR NVL72",
"label": "VR NVL72"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [
"2025",
"2",
"72",
"144"
],
"temporal_mentions": [
"2025"
]
},
"value": {
"numeric_mentions": [
"2025",
"2",
"72",
"144"
],
"value_text": "VR NVL72 最初被稱為 VR NVL144，因為 GTC 2025 的命名方式以系統中的 GPU compute die 數量計算；每個 package 有 2 個 GPU compute dies，Oberon rack 有 72 個 Rubin packages，因此共 144 個 compute dies。"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| entity | VR NVL72 | 04_knowledge_base/VR NVL72 |
| context_0 | VR NVL144 | 04_knowledge_base/VR NVL144 |
| context_1 | GPU | GPU |
| context_2 | Oberon | 04_knowledge_base/NVIDIA Oberon |
| context_3 | Rubin | 04_knowledge_base/Vera Rubin |
⑤ Human Review
請在 Properties 逐項確認：
- 原文 → Atomic Claim 是否忠實
- Atomic Claim → Semantic Frame 是否忠實
- Canonical Entity mapping 是否正確
- Epistemic mode 是否保留原文語氣
- 最後選擇
review_action
Review state
Markdown 內文不是正式 approval。只有 Apply bridge 寫入的 Decision Ledger event 才是正式決策。