clm_9148c3475e93426b8511928038b5755c_r1

IX2-0160

① SA Source

• Source: 開啟完整 SA 文章
• Section: Nvidia versus AMD Disagg Prefill
• Line hint: 266

Context Before

Source: SemiAnalysis InferenceX ↗

Evidence

However, if we were to only measure output (decode) token throughput, we see that output token throughput is much higher for the B200 than for the MI355X at lower interactivity levels

Context After

SemiAnalysis is free open source software and reader-supported. To receive new posts and support our work, consider becoming a free or paid subscriber.

Subscribed

② Atomic Claim

但若只衡量 output（decode）token throughput，在較低 interactivity 下，B200 明顯高於 MI355X。

• Epistemic Mode: ASSERTED
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "comparison_expression": "但若只衡量 output（decode）token throughput，在較低 interactivity 下，B200 明顯高於 MI355X。",
  "entities": [
    {
      "id": "04_knowledge_base/Decode",
      "label": "decode"
    },
    {
      "id": "04_knowledge_base/NVIDIA B200",
      "label": "B200"
    },
    {
      "id": "04_knowledge_base/MI355X",
      "label": "MI355X"
    }
  ],
  "frame_type": "COMPARISON",
  "metric": "THROUGHPUT",
  "operator": "GREATER_THAN",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [],
    "temporal_mentions": []
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
comparison_entity_0	decode	Decode
comparison_entity_1	B200	04_knowledge_base/NVIDIA B200
comparison_entity_2	MI355X	MI355X

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。