{
  "source_file": "0930",
  "source_platform": "Dev.to",
  "author": "Julien Simon",
  "title_listed": "Video — Deep Dive: Optimizing LLM inference",
  "keywords": "",
  "domain": "大模型算法",
  "content_type": "深度技术解析",
  "is_paid_flag": "否",
  "url": "https://dev.to/juliensimon/video-deep-dive-optimizing-llm-inference-1g31",
  "publish_date": "2024-03-11",
  "corpus_stage": "",
  "notes": "",
  "id": "23dbac39bb78",
  "collected_at": "2026-10-02T14:01:05",
  "title": "Video — Deep Dive: Optimizing LLM inference - DEV Community",
  "body_markdown": "Video — Deep Dive: Optimizing LLM inference\nOpen-source LLMs are great for conversational applications, but they can be difficult to scale in production and often deliver latency and throughput that are incompatible with your cost-performance objectives.\nIn this video, we zoom in on optimizing LLM inference, and study key mechanisms that help reduce latency and increase throughput: the KV cache, continuous batching, and speculative decoding, including the state-of-the-art Medusa approach.",
  "metadata": {
    "title": "Video — Deep Dive: Optimizing LLM inference - DEV Community",
    "description": "Video — Deep Dive: Optimizing LLM inference     Open-source LLMs are great for... Tagged with llm, deeplearning, transformers, mlops.",
    "keywords": "llm, deeplearning, transformers, mlops, software, coding, development, engineering, inclusive, community",
    "author": null,
    "og:type": "article",
    "og:url": "https://dev.to/juliensimon/video-deep-dive-optimizing-llm-inference-1g31",
    "og:title": "Video — Deep Dive: Optimizing LLM inference",
    "og:description": "Video — Deep Dive: Optimizing LLM inference     Open-source LLMs are great for...",
    "og:site_name": "DEV Community",
    "og:image": "https://media2.dev.to/dynamic/image/width=1000,height=500,fit=cover,gravity=auto,format=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fcxjb2aq79843d3xutieq.png",
    "twitter:site": "@thepracticaldev",
    "twitter:creator": "@julsimon",
    "twitter:title": "Video — Deep Dive: Optimizing LLM inference",
    "twitter:description": "Video — Deep Dive: Optimizing LLM inference     Open-source LLMs are great for...",
    "twitter:card": "summary_large_image",
    "twitter:widgets:new-embed-design": "on",
    "twitter:image:src": "https://media2.dev.to/dynamic/image/width=1000,height=500,fit=cover,gravity=auto,format=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fcxjb2aq79843d3xutieq.png"
  },
  "paywall": {
    "listed_paid": false,
    "detected": false,
    "marker": "",
    "public_only": false,
    "policy": "付费/会员内容只保留页面公开可见部分，不绕过付费墙"
  },
  "crawl": {
    "success": true,
    "status_code": 200,
    "error": ""
  },
  "code_blocks": [],
  "images": [
    {
      "status": "http_404",
      "url": "https://res.cloudinary.com/practicaldev/image/fetch/s--TARvgIPr--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://cdn-images-1.medium.com/max/1024/1%2AJj8FxGzz1DLpdZqr8xf5Pg.png",
      "src": "https://res.cloudinary.com/practicaldev/image/fetch/s--TARvgIPr--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://cdn-images-1.medium.com/max/1024/1%2AJj8FxGzz1DLpdZqr8xf5Pg.png",
      "alt": "",
      "downloaded": false,
      "retried_at": "2026-10-20T17:58:21Z"
    }
  ],
  "audio": [],
  "videos": [
    {
      "url": "https://www.youtube.com/watch?v=hMs8VNRy5Ys",
      "src": "https://www.youtube.com/embed/hMs8VNRy5Ys",
      "status": "ok",
      "kind": "single_video",
      "local_path": "attachments/23dbac39bb78/videos/repair-hMs8VNRy5Ys.mp4",
      "bytes": 139854403,
      "valid": true,
      "height": 1080,
      "width": 1920,
      "duration": "2171.081000",
      "reused": false
    }
  ],
  "links": [
    {
      "href": "https://res.cloudinary.com/practicaldev/image/fetch/s--TARvgIPr--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://cdn-images-1.medium.com/max/1024/1%2AJj8FxGzz1DLpdZqr8xf5Pg.png",
      "label": "",
      "type": "a"
    },
    {
      "href": "https://www.youtube.com/embed/hMs8VNRy5Ys",
      "label": "",
      "type": "iframe"
    }
  ],
  "repair_version": 2,
  "repaired_at": "2026-10-14T10:49:06Z",
  "source_url": "https://dev.to/juliensimon/video-deep-dive-optimizing-llm-inference-1g31",
  "papers": [],
  "actual_video_files": [],
  "body_chars": 493,
  "expected_body_images": 1,
  "discovery_status": "ok",
  "delivery_status": "partial",
  "missing_reason": "http_404",
  "raw_html": "attachments/23dbac39bb78/pages/source.html",
  "local_html": "attachments/23dbac39bb78/pages/article.html",
  "body_base_path": "attachments/23dbac39bb78/pages/",
  "body_markdown_path": "attachments/23dbac39bb78/pages/article.md"
}