clm_79a0d7df36b6446e80f035e9dc8467d5_r1

IX2-0296

① SA Source

• Source: 開啟完整 SA 文章
• Section: AMD ATOM Engine
• Line hint: 494

Context Before

Source: SemiAnalysis InferenceX ↗

AMD ATOM Engine

Evidence

AMD has launched a new inference engine called ATOM. Atom can deliver slightly better single node performance, but it is completely lacking on a lot of features that makes it unusable for real workloads. One such example is that it does not support NVMe or CPU KVCache offloading, tool parsing, wide expert parallelism, or disaggregated serving. This has led to zero customers using it in production. Unlike Nvidia’s TRTLLM which generates billions of tokens per hour globally at companies like TogetherAI, etc and does support tool parsing and other features ↗, there are no token factories currently using ATOM due to the lack of the aforementioned features.

Context After

Furthermore, maintainers of open-source inference engines like vLLM are disappointed in AMD due to a lack of engineering and GPU resources provided by AMD. For example, Simon Mo, lead vLLM maintainer, states in this GitHub RFC that there is still no working MI355X that he can add to vLLM CI, hence the poor user experience. There are currently zero Mi355X tests on vLLM, while NVIDIA’s B200 has many tests on vLLM. Similarly, there are still not enough MI300X CI machines on vLLM. Upstream vLLM needs at least 20 more MI300 machines, 20 more MI325 machines and 20 more MI355X machines to reach the same level of usability as CUDA.

We at SemiAnalysis have been trying to get AMD to contribute more compute to vLLM and have had some success on that within the couple weeks. vLLM will start to get a couple of MI355X machines such that they can bring their CI test parity from 0% to non-0%. We will talk more about AMD’s previous lackluster contribution towards vLLM, SGLang, PyTorch CI machine situation & how Anush started to fix it in our upcoming State of AMD article. At SemiAnalysis, we will have internal dashboard to track the # of tests & quality of tests that AMD & NVIDIA runs on vLLM, SGLang, PyTorch, & JAX.

② Atomic Claim

相較之下，Nvidia 的 TRTLLM 支援 tool parsing 等功能，並已在 TogetherAI 等公司全球每小時產生數十億 tokens；ATOM 因缺少前述功能，目前沒有 token factory 使用。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "comparison_expression": "相較之下，Nvidia 的 TRTLLM 支援 tool parsing 等功能，並已在 TogetherAI 等公司全球每小時產生數十億 tokens；ATOM 因缺少前述功能，目前沒有 token factory 使用。",
  "entities": [
    {
      "id": "02_companies/NVDA",
      "label": "Nvidia"
    },
    {
      "id": "04_knowledge_base/TensorRT-LLM",
      "label": "TRTLLM"
    },
    {
      "id": "04_knowledge_base/AMD ATOM inference engine",
      "label": "ATOM"
    }
  ],
  "frame_type": "COMPARISON",
  "metric": "UNSPECIFIED_METRIC",
  "operator": "COMPARES_WITH",
  "qualifiers": {
    "condition_text": null,
    "numeric_mentions": [],
    "temporal_mentions": []
  }
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
comparison_entity_0	Nvidia	NVDA
comparison_entity_1	TRTLLM	TensorRT-LLM
comparison_entity_2	ATOM	04_knowledge_base/AMD ATOM inference engine

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。