ai-engineering-from-scratch-zh/phases/17-infrastructure-and-production/01-managed-llm-platforms/quiz.json at main · fancyboi999/ai-engineering-from-scratch-zh · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
{
  "lesson": "01-managed-llm-platforms",
  "title": "托管 LLM 平台 —— Bedrock、Vertex AI、Azure OpenAI",
  "questions": [
    {
      "stage": "pre",
      "question": "在学习本课之前，如果一个客户需要把 Claude、Llama 和 Cohere 放在同一个 API 之后使用，你会首先考虑哪个平台？",
      "options": [
        "Vertex AI",
        "AWS Bedrock",
        "Azure OpenAI Service",
        "直接使用 Anthropic API"
      ],
      "correct": 1,
      "explanation": ""
    },
    {
      "stage": "pre",
      "question": "哪家超大规模云厂商的押注被描述为独家合作关系，而非市场（marketplace）模式？",
      "options": [
        "Azure OpenAI",
        "Vertex AI",
        "AWS Bedrock",
        "OCI Generative AI"
      ],
      "correct": 0,
      "explanation": ""
    },
    {
      "stage": "check",
      "question": "在 Llama 3.1 405B 同等模型上，Azure OpenAI（使用 PTU）与 Bedrock 按需（on-demand）之间实测的 TTFT 中位数差距大约是多少？",
      "options": [
        "约 100 ms",
        "约 5 ms",
        "约 250 ms",
        "约 25 ms"
      ],
      "correct": 3,
      "explanation": ""
    },
    {
      "stage": "check",
      "question": "Bedrock 在 2025 年的哪个特性能原生地在 CloudWatch 中提供最清晰的按产品成本归因？",
      "options": [
        "Bedrock Knowledge Bases",
        "Application Inference Profiles",
        "Bedrock Guardrails",
        "Bedrock Agents"
      ],
      "correct": 1,
      "explanation": ""
    },
    {
      "stage": "check",
      "question": "Azure PTU 通常在多高的持续利用率区间上与按需模式打平（break even）？",
      "options": [
        "5-10%",
        "80-95%",
        "20-30%",
        "40-60%"
      ],
      "correct": 3,
      "explanation": ""
    },
    {
      "stage": "post",
      "question": "为什么本课推荐的 2026 年策略是：对产品关键路径的 LLM 调用至少采用两家供应商？",
      "options": [
        "这是 SOC 2 Type II 的强制要求",
        "前沿模型的领先地位每月轮换，因此单一供应商锁定会让你与三分之二的前沿模型无缘",
        "超大规模云厂商除非你同时使用其竞争对手，否则拒绝签署 BAA",
        "单一供应商的定价总是更贵"
      ],
      "correct": 1,
      "explanation": ""
    },
    {
      "stage": "post",
      "question": "哪一项表述最准确地描述了这三个平台的 FinOps（财务运营）状况？",
      "options": [
        "Azure 原生能力最清晰，Bedrock 最不透明",
        "Vertex 完全没有任何成本归因能力",
        "三者都暴露了完全相同的按请求归因能力",
        "Bedrock 原生能力最清晰，Vertex 借助 BigQuery 最灵活，Azure 在缺乏埋点时最不透明"
      ],
      "correct": 3,
      "explanation": ""
    }
  ]
}