Đồ Họa Thông Tin So Sánh Kiến Trúc Mô Hình

{
  "type": "side-by-side AI architecture comparison infographic",
  "style": "clean technical diagram, white background, thin black outlines, rounded rectangles, dashed callout boxes, color-coded highlights, presentation-slide aesthetic, vector infographic",
  "canvas": {
    "aspect_ratio": "2:1",
    "resolution": "wide horizontal"
  },
  "title_row": {
    "left_title": "[DeepSeek V3/R1 (671 billion)]",
    "right_title": "[DeepSeek V4 (1.2 trillion)]",
    "left_title_color": "bright orange-red",
    "right_title_color": "bright blue"
  },
  "layout": {
    "columns": 2,
    "sections": [
      {
        "title": "[DeepSeek V3/R1 (671 billion)]",
        "position": "left half",
        "count": 9,
        "labels": [
          "Vocabulary size of 129k",
          "FeedForward (SwiGLU) module",
          "Intermediate hidden layer dimension of 2,048",
          "MoE layer",
          "Supported context length of 128k tokens",
          "First 3 blocks use dense FFN with hidden size 18,432 instead of MoE",
          "Sample input text",
          "Embedding dimension of 7,168",
          "128 heads"
        ]
      },
      {
        "title": "[DeepSeek V4 (1.2 trillion)]",
        "position": "right half",
        "count": 9,
        "labels": [
          "Vocabulary size of 160k",
          "FeedForward (SwiGLU) module",
          "Intermediate hidden layer dimension of 3,072",
          "MoE layer",
          "Supported context length of 256k tokens",
          "First 3 blocks use dense FFN with hidden size 24,576 instead of MoE",
          "Sample input text",
          "Embedding dimension of 8,192",
          "128 heads"
        ]
      },
      {
        "title": "bottom comparison table",
        "position": "bottom full width",
        "count": 10,
        "labels": [
          "Total parameters",
          "Active parameters per token",
          "Hidden size",
          "Intermediate (FF)",
          "Attention heads",
          "Context length",
          "Embedding dimension",
          "Vocabulary size"
        ]
      }
    ]
  },
  "left_panel": {
    "background": "very light gray rounded rectangle",
    "main_stack": {
      "count": 8,
      "blocks": [
        "Tokenized text",
        "Token embedding layer",
        "RMSNorm 1",
        "Multi-head Latent Attention",
        "RMSNorm 2",
        "MoE",
        "Final RMSNorm",
        "Linear output layer"
      ]
    },
    "side_module": "RoPE attached to the attention block on the left side",
    "attention_block": {
      "label": "Multi-head Latent Attention",
      "accent": "orange-red text for the word Latent"
    },
    "feedforward_inset": {
      "title": "FeedForward (SwiGLU) module",
      "count": 4,
      "blocks": [
        "Linear layer",
        "SiLU activation",
        "Linear layer",
        "Linear layer"
      ],
      "diagram": "two branches multiplied, then projected"
    },
    "moe_inset": {
      "title": "MoE layer",
      "count": 5,
      "blocks": [
        "top combine node",
        "Feed forward",
        "Feed forward",
        "Router",
        "expert count badge 256"
      ],
      "details": "small black square with 1 selected expert, arrows routing upward to experts, dotted divider line"
    },
    "annotations": {
      "vocab": "Vocabulary size of 129k",
      "ff_dim": "Intermediate hidden layer dimension of 2,048",
      "context": "Supported context length of 128k tokens",
      "dense_first_blocks": "First 3 blocks use dense FFN with hidden size 18,432 instead of MoE",
      "resource_savings": "Resource savings: Model size is 671B but only 1 (shared) + 8 experts active per token; only 37B parameters are active per inference step"
    },
    "bottom_stats": {
      "count": 8,
      "items": [
        "Total parameters: 671B",
        "Active parameters per token: 37B (1 + 8 experts)",
        "Hidden size: 7,168",
        "Intermediate (FF): 2,048",
        "Attention heads: 128",
        "Context length: 128k",
        "Embedding dimension: 7,168",
        "Vocabulary size: 129k"
      ]
    }
  },
  "right_panel": {
    "background": "very light blue rounded rectangle",
    "main_stack": {
      "count": 8,
      "blocks": [
        "Tokenized text",
        "Token embedding layer",
        "RMSNorm 1",
        "Multi-head Latent Attention",
        "RMSNorm 2",
        "MoE",
        "Final RMSNorm",
        "Linear output layer"
      ]
    },
    "side_module": "RoPE attached to the attention block on the left side",
    "attention_block": {
      "label": "Multi-head Latent Attention",
      "accent": "blue text for the word Latent"
    },
    "feedforward_inset": {
      "title": "FeedForward (SwiGLU) module",
      "count": 4,
      "blocks": [
        "Linear layer",
        "SiLU activation",
        "Linear layer",
        "Linear layer"
      ],
      "diagram": "same structure as left panel"
    },
    "moe_inset": {
      "title": "MoE layer",
      "count": 5,
      "blocks": [
        "top combine node",
        "Feed forward",
        "Feed forward",
        "Router",
        "expert count badge 384"
      ],
      "details": "small black square with 1 selected expert, arrows routing upward to experts, dotted divider line, blue border emphasis"
    },
    "annotations": {
      "vocab": "Vocabulary size of 160k",
      "ff_dim": "Intermediate hidden layer dimension of 3,072",
      "context": "Supported context length of 256k tokens",
      "dense_first_blocks": "First 3 blocks use dense FFN with hidden size 24,576 instead of MoE",
      "resource_savings": "Resource savings: Model size is 1.2T but only 1 (shared) + 8 experts active per token; only 52B parameters are active per inference step"
    },
    "bottom_stats": {
      "count": 8,
      "items": [
        "Total parameters: 1.2T",
        "Active parameters per token: 52B (1 + 8 experts)",
        "Hidden size: 8,192",
        "Intermediate (FF): 3,072",
        "Attention heads: 128",
        "Context length: 256k",
        "Embedding dimension: 8,192",
        "Vocabulary size: 160k"
      ]
    }
  },
  "global_notes": "Create a highly detailed transformer architecture comparison diagram with mirrored layouts. Each half contains one large model stack diagram plus 2 inset diagrams: 1 feedforward module and 1 MoE layer. Use arrows between blocks, tiny technical labels, and connector lines from labels to the relevant components. Keep the typography dense and slide-like, with orange-red used for left-side emphasis and blue used for right-side emphasis. Include a bottom row of compact tabular metrics spanning the width."
}