clm_55ce8f02ebc44d45b986027dc88720db_r1

2025-06-13_amd-advancing-ai-mi350x-and-mi400-ualoe72-mi500-ual256::AMD25-0004

① SA Source

• Source: 開啟完整 SA 文章
• Section: Executive Summary
• Line hint: 33

Context Before

The MI355X is competitive with the HGX B200 for small to medium model inferencing but it will not be competitive against the GB200 NVL72

Evidence

Despite AMD’s marketing RDF ↗, the MI355 128 GPU rack is not a “rack scale solution” – it only has a scale up world size of 8 GPUs versus the GB200 NVL72 which has a world size of 72 GPUs.

Context After

The MI355X will have similar collective performance as the HGX B200 but MI355X collectives will run at least 18x slower than on the GB200 NVL72, if not even slower

② Atomic Claim

MI355 128-GPU rack 的 scale-up world size 只有 8 GPUs；GB200 NVL72 為 72 GPUs。

• Epistemic Mode: ASSERTED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "comparison_expression": "MI355: 8 GPUs; GB200 NVL72: 72 GPUs",
  "entities": [
    {
      "id": "04_knowledge_base/MI355",
      "label": "MI355"
    },
    {
      "id": "04_knowledge_base/GB200 NVL72",
      "label": "GB200 NVL72"
    },
    {
      "id": "04_knowledge_base/GPU",
      "label": "GPU"
    }
  ],
  "frame_type": "COMPARISON",
  "metric": "scale_up_world_size",
  "operator": "8_VS_72",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [
      "128",
      "8",
      "72"
    ],
    "temporal_mentions": []
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
comparison_entity_0	MI355	MI355
comparison_entity_1	GB200 NVL72	04_knowledge_base/GB200 NVL72
comparison_entity_2	GPU	GPU

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。