{
  "docId": "019dd923-5e88-73ef-bd5d-0856e1444fb9",
  "docSlug": "8757f1b44ef7f176",
  "documentTitle": "2024 Air Street Capital The State of AI Report 2024",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 24,
  "pageCount": 213,
  "prevPage": 23,
  "nextPage": 25,
  "slideType": "industry_trends",
  "function": "summarize",
  "density": "dense",
  "nDataPoints": 6,
  "notes": "Includes a screenshot of a chat interface demonstrating Phi-3-mini-4k-instruct-q4 capabilities.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "screenshot"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-0856e1444fb9/24",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-0856e1444fb9",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-0856e1444fb9.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-0856e1444fb9#slide-24",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "As big tech companies think through large-scale end user deployment, we're starting to see high-performing LLM and multimodal models that are small enough to run on smartphones.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c3-7407-8d0f-76fe245339a9",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.6,
        "w": 0.26,
        "x": 0.72,
        "y": 0.28
      },
      "kind": "image",
      "text": "Phi-3-mini-4k-instruct-q4 chat interface",
      "attrs": null,
      "subkind": "screenshot",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "b687b764-351b-41a5-9ff1-392205c6e1c7",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.55,
        "w": 0.68,
        "x": 0.02,
        "y": 0.31
      },
      "kind": "list",
      "text": "Microsoft's phi-3.5-mini is a 3.8B LM that competes with larger models like 7B and Llama 3.1 8B. It performs well on reasoning and question-answering, but size restricts its factual knowledge. To enable on-device inference, the model was quantized to 4 bits, reducing its memory footprint to approximately 1.8GB.\nApple introduced MobileCLIP, a family of efficient image-text models optimized for fast inference on smartphones. Using novel multimodal reinforced training, they improve the accuracy of compact models by transferring knowledge from an image captioning model and an ensemble of strong CLIP encoders.\nHugging Face also got in on the action with SmolLM, a family of small language models, available in 135M, 360M, and 1.7B formats. By using a highly curated synthetic dataset created via an enhanced version of Cosmopedia (see slide 31) the team achieved SOTA performance for the size.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "5dec3ea8-8ad3-451d-9ad9-f58392811ba8",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.68,
        "x": 0.02,
        "y": 0.21
      },
      "kind": "paragraph",
      "text": "As big tech companies think through large-scale end user deployment, we're starting to see high-performing LLM and multimodal models that are small enough to run on smartphones.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "89321bd6-5625-4766-b74f-b0dc91c724e1",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.04,
        "w": 0.5,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "Models built for mobile compete with their larger peers",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "738bcec3-6a70-4528-a49d-32b1016e6bff",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Action Titles",
      "slug": "action-titles",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a958-942f41cc52a9",
      "evidence": "Title 'Models built for mobile compete with their larger peers'",
      "confidence": 85
    },
    {
      "name": "Audience Definition",
      "slug": "audience-definition",
      "agent": "Storyteller",
      "layer": "slide",
      "matchId": "3451a0e7-a8f2-407d-bf02-ab03d4b874d0",
      "evidence": "As big tech companies think through large-scale end user deployment, we're starting to see high-performing LLM and multimodal models that are small enough to run on smartphones.",
      "confidence": 0.5
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 153,
      "from": 9,
      "beatId": "019dd95a-0682-776c-8e35-5f2398b8d1d0",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Exec summary + Research + Industry sections inventory the year",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 85,
      "from": 12,
      "beatId": "019dd95a-0682-776c-8e35-6d9d8ee06d0a",
      "arcName": "Voyage and Return",
      "arcSlug": "voyage-return",
      "beatName": "The Unknown",
      "beatSlug": "voyage-return-the-unknown",
      "evidence": "Research section explores frontier model uncharted territory",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Complication",
      "parentBeatSlug": "complication"
    }
  ],
  "loops": [
    {
      "to": 29,
      "from": 21,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3e-2b2229583eee",
      "evidence": "Eight slides on shrinking models, on-device, hybrids, transformers reigning",
      "position": 3,
      "objective": "Demonstrate efficiency frontier across distillation, quantization, hybrids",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 75,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}