IX2-0088
① SA Source
- Source: 開啟完整 SA 文章
- Section:
Tracking Improvements Over Time - Line hint:
117
Context Before
TP allows for maximize interactivity at small batch sizes, but it must carry out an all-reduce at every layer. EP shards experts, exploiting MoE sparsity, with the drawback being an all-to-all collective (which is more costly than simpler collectives like all-reduce) is carried out for MoE layers and can be imbalanced at small batches. DP replicates the entire model (or just parts of a model, like attention) on multiple groups of GPUs (ranks) and then load balances requests among ranks. It is the simplest to scale, but repeats weight loading which can be wasteful at scale.
Tracking Improvements Over Time
Evidence
While new chips are released on an O(yearly) cadence, software releases happen on an O(weekly) cadence
Context After
DeepSeek R1
The AMD team has significantly improved performance for all configurations of SGLang DeepSeek R1 FP4. For the same interactivity, AMD has almost doubled the amount of throughput in the span of less than 2 months. Moreover, we have pushed AMD to upstream performance enhancing changes from their forked SGLang images into the official SGLang image. From December 2025 to January 2026, AMD’s software was improved up to 2x in performance.
② Atomic Claim
關於 InferenceX：新晶片大約以年度 cadence 推出，但 software releases 大約是每週 cadence。
- Epistemic Mode:
ASSERTED - Mapping Status:
COMPLETE
③ Semantic Frame
{
"attribute": "SOFTWARE_STATUS",
"context_nodes": [],
"entity": {
"id": "04_knowledge_base/InferenceX",
"label": "InferenceX"
},
"frame_type": "ATTRIBUTE",
"qualifiers": {
"condition_text": null,
"numeric_mentions": [],
"temporal_mentions": []
},
"value": {
"numeric_mentions": [],
"value_text": "關於 InferenceX：新晶片大約以年度 cadence 推出，但 software releases 大約是每週 cadence。"
}
}④ Canonical Entity Mapping
| Role | Surface Label | Canonical Target |
|---|---|---|
| entity | InferenceX | InferenceX |
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。