https://vast.ai

{
  "url": "https://vast.ai",
  "timestamp": 1776792074480,
  "fetch": {
    "mode": "simple",
    "timeMs": 110,
    "htmlSizeBytes": 79818,
    "supportsMarkdown": false,
    "markdownAgents": {
      "contentNegotiation": false,
      "mdUrl": {
        "found": false,
        "url": null
      },
      "linkTag": {
        "found": false,
        "url": null
      },
      "linkHeader": {
        "found": false,
        "url": null
      },
      "responseHeaders": {
        "contentSignal": null,
        "xMarkdownTokens": null,
        "vary": null
      },
      "frontmatter": {
        "present": false,
        "fields": [],
        "level": "none"
      },
      "level": "none"
    },
    "statusCode": 200
  },
  "extraction": {
    "title": "GPU Infrastructure for AI Agents | Vast.ai",
    "excerpt": "Rent high-performance cloud GPUs at low cost with Vast.ai. Instantly deploy GPU rentals for AI, machine learning, deep learning, and rendering. Flexible pricing, fast setup, and global availability",
    "byline": null,
    "siteName": "Vast AI",
    "lang": "en",
    "contentLength": 1508,
    "metadata": {
      "description": "Rent high-performance cloud GPUs at low cost with Vast.ai. Instantly deploy GPU rentals for AI, machine learning, deep learning, and rendering. Flexible pricing, fast setup, and global availability",
      "ogTitle": "GPU Infrastructure for AI Agents | Vast.ai",
      "ogDescription": "Rent high-performance cloud GPUs at low cost with Vast.ai. Instantly deploy GPU rentals for AI, machine learning, deep learning, and rendering. Flexible pricing, fast setup, and global availability",
      "ogImage": "https://vast.ai/home/Vast_Social_Banner.png",
      "ogType": null,
      "canonical": "https://vast.ai/",
      "lang": "en",
      "schemas": [
        {
          "@context": "https://schema.org",
          "@type": "Organization",
          "name": "Vast.ai",
          "url": "https://vast.ai",
          "logo": "https://vast.ai/home/Vast_Social_Banner.png",
          "foundingDate": "2018",
          "description": "AI compute platform connecting developers with high-performance GPU cloud resources.",
          "sameAs": [
            "https://discord.gg/vast",
            "https://github.com/vast-ai",
            "https://www.youtube.com/@vast_ai/videos",
            "https://x.com/vast_ai"
          ]
        },
        {
          "@context": "https://schema.org",
          "@type": "WebSite",
          "name": "Vast.ai",
          "url": "https://vast.ai",
          "potentialAction": {
            "@type": "SearchAction",
            "target": {
              "@type": "EntryPoint",
              "urlTemplate": "https://cloud.vast.ai/create/?query={search_term_string}"
            },
            "query-input": "required name=search_term_string"
          }
        }
      ],
      "robotsMeta": null,
      "author": null,
      "generator": null,
      "markdownAlternateHref": null
    }
  },
  "markdown": "## Real-time GPU infrastructure\n\nPrices set by supply and demand across 20,000+ GPUs. Transparent. Programmatically queryable.\n\n## How it works\n\nFrom sign-up to running GPU workloads in under five minutes.\n\n1\n\n### Add credit & get your API key\n\nStart with as little as $5. Grab your API key from the console — no contracts, no sales calls.\n\n2\n\n### Search GPUs\n\nFilter by model, VRAM, price, and availability — via console or API.\n\n3\n\n### Deploy\n\nLaunch instances in seconds. Scale up or down programmatically.\n\n## Compare. Launch. Exit. Repeat.\n\nEvery GPU on Vast.ai is provisioned through code. The same API that developers use to deploy in seconds is the interface agents use to procure and optimize at scale.\n\nProgrammatic compute provisioning in five lines of code.\n\n`pip install vastai-sdk`\n\nSearch, filter, and deploy from your terminal.\n\n`pip install vastai`\n\nThe interface agents call to provision infrastructure.\n\n`curl -H \"Authorization: Bearer $VAST_API_KEY\" https://cloud.vast.ai/api/v1/bundles/`\n\n## One platform. Three ways to deploy.\n\nGPU Cloud for full control. Serverless for zero-ops inference. Clusters for large-scale training.\n\n## Popular Models, Ready to Deploy\n\nLaunch pre-configured templates for the most popular open-source models.\n\n> “Vast.ai reduced our GPU costs by over 60% while giving us the flexibility to scale training jobs on demand. We serve 200K daily users without breaking the bank.”\n\n## How teams build on Vast.ai\n\nSee how teams use Vast.ai to scale AI infrastructure and accelerate production workloads.\n\n[\n\n![Creatix Technology](https://vast.ai/_next/image?url=%2Fcase-studies%2Fimg-caseStudy-creatix-sm.webp&w=2048&q=75)\n\n](https://vast.ai/case-studies/creatix-technology)\n\n[\n\n![PAICON](https://vast.ai/_next/image?url=%2Fcase-studies%2Fimg-caseStudy-paicon-sm.webp&w=2048&q=75)\n\n](https://vast.ai/case-studies/paicon)\n\n## Start with $5. Scale to 20,000 GPUs.\n\nNo humans required.\n",
  "fullPageMarkdown": "GPU Infrastructure for AI Agents | Vast.ai\n\n[![Vast AI](https://vast.ai/vast_logo.svg)](https://vast.ai/)\n\n[Contact Sales](https://vast.ai/contact-sales)[Console](https://cloud.vast.ai/)\n\n-   Developers\n\n    [Quickstart](https://docs.vast.ai/cli/get-started)[CLI](https://vast.ai/developers/cli)[Python SDK](https://vast.ai/developers/sdk)[API](https://vast.ai/developers/api)[Docs](https://docs.vast.ai/)\n\n-   [Pricing](https://vast.ai/pricing)\n-   Products\n\n    [GPU Cloud](https://vast.ai/products/gpu-cloud)[Clusters](https://vast.ai/products/clusters)[Serverless](https://vast.ai/products/serverless)[Model Library](https://vast.ai/model-library)\n\n-   Hosting\n\n    [Hosting](https://vast.ai/hosting)[Data Centers](https://vast.ai/data-center-application)[Financing](https://vast.ai/hosting/finance)[Hardware](https://vast.ai/hosting/hardware)[Docs](https://docs.vast.ai/documentation/host/hosting-overview)\n\n-   Use Cases\n\n    [All Use Cases](https://vast.ai/use-cases)[AI Agents](https://vast.ai/use-cases/ai-agents)[AI Fine Tuning](https://vast.ai/use-cases/ai-fine-tuning)[AI Image + Video Generation](https://vast.ai/use-cases/ai-image-video-generation)[AI Text Generation](https://vast.ai/use-cases/ai-text-generation)[AI/ML Frameworks](https://vast.ai/use-cases/ai-ml-frameworks)[Audio-to-Text Transcription](https://vast.ai/use-cases/audio-to-text-transcription)[Batch Data Processing](https://vast.ai/use-cases/batch-data-processing)[GPU Programming](https://vast.ai/use-cases/gpu-programming)[Graphics Rendering](https://vast.ai/use-cases/3d-rendering)[Virtual Computing](https://vast.ai/use-cases/virtual-computing)\n\n-   Company\n\n    [About](https://vast.ai/about)[Blog](https://vast.ai/blog)[Careers](https://vast.ai/careers)[Enterprise](https://vast.ai/enterprise)[Case Studies](https://vast.ai/case-studies)[Startup Program](https://vast.ai/startup)[FAQ](https://docs.vast.ai/faq)[Press Releases](https://vast.ai/press-releases)\n\nSOC 2 Certified\n\n# Agent-Ready AI\nInfrastructure.\n\nVast is the infrastructure layer where AI agents autonomously design, procure, and optimize their own compute. API-native provisioning. Real-time pricing. Per-second billing.\n\nGet Started\n\n[CLI Docs](https://docs.vast.ai/cli/get-started)\n\n700K+ transactions/mo20,000+ GPUs40+ data centers68+ GPU types\n\nTrusted by developers and AI teams worldwide\n\n![CHAI](https://vast.ai/partners/Chai.svg)\n\n![BOSCH](https://vast.ai/partners/bosch.svg)\n\n![Cognition](https://vast.ai/partners/Path%201.svg)\n\n![Inria](https://vast.ai/partners/inria.svg)\n\n![IBM](https://vast.ai/partners/ibm-svgrepo-com.svg)\n\n![Brave](https://vast.ai/partners/brave.svg)\n\n![Speechify](https://vast.ai/partners/speechify.svg)\n\n![CHAI](https://vast.ai/partners/Chai.svg)\n\n![BOSCH](https://vast.ai/partners/bosch.svg)\n\n![Cognition](https://vast.ai/partners/Path%201.svg)\n\n![Inria](https://vast.ai/partners/inria.svg)\n\n![IBM](https://vast.ai/partners/ibm-svgrepo-com.svg)\n\n![Brave](https://vast.ai/partners/brave.svg)\n\n![Speechify](https://vast.ai/partners/speechify.svg)\n\n## Real-time GPU infrastructure\n\nPrices set by supply and demand across 20,000+ GPUs. Transparent. Programmatically queryable.\n\n[View All GPUs](https://vast.ai/pricing)\n\n## How it works\n\nFrom sign-up to running GPU workloads in under five minutes.\n\n1\n\n### Add credit & get your API key\n\nStart with as little as $5. Grab your API key from the console — no contracts, no sales calls.\n\n2\n\n### Search GPUs\n\nFilter by model, VRAM, price, and availability — via console or API.\n\n3\n\n### Deploy\n\nLaunch instances in seconds. Scale up or down programmatically.\n\n[Get Started for $5](https://cloud.vast.ai/)\n\n## Compare. Launch. Exit. Repeat.\n\nEvery GPU on Vast.ai is provisioned through code. The same API that developers use to deploy in seconds is the interface agents use to procure and optimize at scale.\n\nPython SDK[Docs →](https://docs.vast.ai/sdk/python/quickstart)\n\nProgrammatic compute provisioning in five lines of code.\n\n`pip install vastai-sdk`\n\nCLI[Docs →](https://docs.vast.ai/cli/get-started)\n\nSearch, filter, and deploy from your terminal.\n\n`pip install vastai`\n\nREST API[Docs →](https://docs.vast.ai/api-reference/introduction)\n\nThe interface agents call to provision infrastructure.\n\n`curl -H \"Authorization: Bearer $VAST_API_KEY\" https://cloud.vast.ai/api/v1/bundles/`\n\n[Explore Developer Tools](https://vast.ai/developers)\n\ndeploy.py\n\n## One platform. Three ways to deploy.\n\nGPU Cloud for full control. Serverless for zero-ops inference. Clusters for large-scale training.\n\n### GPU Cloud\n\nOn-demand instances across 40+ data centers and 20,000+ GPUs. Deploy in seconds via CLI, SDK, or API.\n\n[Explore GPU Cloud](https://vast.ai/products/gpu-cloud)\n\n### Serverless\n\nDeploy models as endpoints with automatic benchmarking and optimization across GPU types. Autoscale to zero, pay only for compute time.\n\n[Try Serverless](https://vast.ai/products/serverless)\n\n### Clusters\n\nDedicated multi-node GPU clusters with InfiniBand networking for large-scale training.\n\n[View Clusters](https://vast.ai/products/clusters)\n\n![Case Studies](https://vast.ai/uploads/aurora/images/img-home-caseStudies.png)\n\nBuilt for Every AI Workload\n\nFrom training to inference, fine-tuning to rendering — run any GPU workload on Vast.\n\n[Use Cases](https://vast.ai/use-cases)\n\n[AI/ML Frameworks](https://vast.ai/use-cases/ai-ml-frameworks)[AI Text Generation](https://vast.ai/use-cases/ai-text-generation)[AI Image + Video Generation](https://vast.ai/use-cases/ai-image-video-generation)[AI Agents](https://vast.ai/use-cases/ai-agents)[Batch Data Processing](https://vast.ai/use-cases/batch-data-processing)[Audio-to-Text Transcription](https://vast.ai/use-cases/audio-to-text-transcription)[AI Fine Tuning](https://vast.ai/use-cases/ai-fine-tuning)[Virtual Computing](https://vast.ai/use-cases/virtual-computing)[GPU Programming](https://vast.ai/use-cases/gpu-programming)[Graphics Rendering](https://vast.ai/use-cases/3d-rendering)\n\n## Popular Models, Ready to Deploy\n\nLaunch pre-configured templates for the most popular open-source models.\n\n![](https://vast.ai/model/logos/google.svg)\n\n### Gemma 4 26B A4B IT\n\nGemma 4 26B A4B MoE vision-language model by Google with 256K context and thinking mode\n\n[Deploy](https://vast.ai/model/gemma-4-26b-a4b-it)\n\n![](https://vast.ai/model/logos/google.svg)\n\n### Gemma 4 31B IT\n\nGemma 4 31B dense vision-language model by Google with 256K context and thinking mode\n\n[Deploy](https://vast.ai/model/gemma-4-31b-it)\n\n![](https://vast.ai/model/logos/lightricks.svg)\n\n### LTX-2.3\n\nLTX-2.3 is a DiT-based audio-video foundation model with improved quality and prompt adherence for synchronized video and audio generation\n\n[Deploy](https://vast.ai/model/ltx-2-3)\n\n![](https://vast.ai/model/logos/alibaba.svg)\n\n### Qwen3.5 397B A17B\n\nEfficient multimodal reasoning model with hybrid DeltaNet-attention architecture\n\n[Deploy](https://vast.ai/model/qwen35-397b-a17b)\n\n[Browse Model Library](https://vast.ai/model-library)\n\n> “Vast.ai reduced our GPU costs by over 60% while giving us the flexibility to scale training jobs on demand. We serve 200K daily users without breaking the bank.”\n\n## How teams build on Vast.ai\n\nSee how teams use Vast.ai to scale AI infrastructure and accelerate production workloads.\n\n[\n\n![Creatix Technology](https://vast.ai/_next/image?url=%2Fcase-studies%2Fimg-caseStudy-creatix-sm.webp&w=2048&q=75)\n\n](https://vast.ai/case-studies/creatix-technology)\n\nCreatix Technology\n\n[### Creatix Technology Scales to 200K Daily Users with Vast.ai's GPU Cloud](https://vast.ai/case-studies/creatix-technology)\n\nHow a fast-growing AI app company cut infrastructure costs by over 60% and powered millions of new users with Vast.ai.\n\nTech\n\n[View Case Study](https://vast.ai/case-studies/creatix-technology)\n\n[\n\n![PAICON](https://vast.ai/_next/image?url=%2Fcase-studies%2Fimg-caseStudy-paicon-sm.webp&w=2048&q=75)\n\n](https://vast.ai/case-studies/paicon)\n\nPAICON\n\n[### PAICON Accelerates Global, Data-Centric Cancer Diagnostics with Vast.ai](https://vast.ai/case-studies/paicon)\n\nHow a global oncology data platform used Vast.ai’s GPU cloud to rapidly iterate on Athena—validating that diversity can matter more than scale—while significantly reducing research-phase training costs.\n\nMedical AI\n\n[View Case Study](https://vast.ai/case-studies/paicon)\n\n## Start with $5. Scale to 20,000 GPUs.\n\nNo humans required.\n\n[Get Started](https://cloud.vast.ai/)[Contact Sales](https://vast.ai/contact-sales)\n",
  "markdownStats": {
    "images": 2,
    "links": 2,
    "tables": 0,
    "codeBlocks": 0,
    "headings": 10
  },
  "tokens": {
    "htmlTokens": 25181,
    "markdownTokens": 477,
    "reduction": 24704,
    "reductionPercent": 98
  },
  "score": {
    "score": 71,
    "grade": "C",
    "dimensions": {
      "semanticHtml": {
        "score": 75,
        "weight": 20,
        "grade": "B",
        "checks": {
          "uses_article_or_main": {
            "score": 100,
            "weight": 20,
            "details": "Has <main>"
          },
          "proper_heading_hierarchy": {
            "score": 100,
            "weight": 25,
            "details": "Clean heading hierarchy"
          },
          "semantic_elements": {
            "score": 25,
            "weight": 20,
            "details": "14 semantic elements, 169 divs (ratio: 8%)"
          },
          "meaningful_alt_texts": {
            "score": 42,
            "weight": 15,
            "details": "10/24 images with meaningful alt text"
          },
          "low_div_nesting": {
            "score": 91,
            "weight": 20,
            "details": "Avg div depth: 5.9, max: 9"
          }
        }
      },
      "contentEfficiency": {
        "score": 55,
        "weight": 25,
        "grade": "D",
        "checks": {
          "token_reduction_ratio": {
            "score": 100,
            "weight": 40,
            "details": "98% token reduction (HTML→Markdown)"
          },
          "content_to_noise_ratio": {
            "score": 0,
            "weight": 30,
            "details": "Content ratio: 1.9% (1508 content chars / 79818 HTML bytes)"
          },
          "minimal_inline_styles": {
            "score": 0,
            "weight": 15,
            "details": "78/640 elements with inline styles (12.2%)"
          },
          "reasonable_page_weight": {
            "score": 100,
            "weight": 15,
            "details": "HTML size: 78KB"
          }
        }
      },
      "aiDiscoverability": {
        "score": 55,
        "weight": 25,
        "grade": "D",
        "checks": {
          "has_llms_txt": {
            "score": 100,
            "weight": 20,
            "details": "llms.txt exists and is valid"
          },
          "has_robots_txt": {
            "score": 100,
            "weight": 10,
            "details": "robots.txt exists"
          },
          "robots_allows_ai_bots": {
            "score": 100,
            "weight": 15,
            "details": "All major AI bots allowed"
          },
          "has_sitemap": {
            "score": 100,
            "weight": 10,
            "details": "Sitemap found"
          },
          "supports_markdown_negotiation": {
            "score": 0,
            "weight": 25,
            "details": "No Markdown for Agents support detected"
          },
          "has_content_signals": {
            "score": 0,
            "weight": 20,
            "details": "No Content-Signal found (robots.txt or HTTP headers)"
          }
        }
      },
      "structuredData": {
        "score": 100,
        "weight": 15,
        "grade": "A",
        "checks": {
          "has_schema_org": {
            "score": 100,
            "weight": 30,
            "details": "JSON-LD found: Organization, WebSite"
          },
          "has_open_graph": {
            "score": 100,
            "weight": 25,
            "details": "All OG tags present"
          },
          "has_meta_description": {
            "score": 100,
            "weight": 20,
            "details": "Meta description: 197 chars"
          },
          "has_canonical_url": {
            "score": 100,
            "weight": 15,
            "details": "Canonical URL present"
          },
          "has_lang_attribute": {
            "score": 100,
            "weight": 10,
            "details": "lang=\"en\""
          }
        }
      },
      "accessibility": {
        "score": 93,
        "weight": 15,
        "grade": "A",
        "checks": {
          "content_without_js": {
            "score": 100,
            "weight": 40,
            "details": "Content available without JavaScript"
          },
          "reasonable_page_size": {
            "score": 100,
            "weight": 30,
            "details": "Page size: 78KB"
          },
          "fast_content_position": {
            "score": 75,
            "weight": 30,
            "details": "Main content starts at 36% of HTML"
          }
        }
      }
    }
  },
  "recommendations": [
    {
      "id": "improve_content_ratio",
      "priority": "critical",
      "category": "contentEfficiency",
      "titleKey": "rec.improve_content_ratio.title",
      "descriptionKey": "rec.improve_content_ratio.description",
      "howToKey": "rec.improve_content_ratio.howto",
      "effort": "moderate",
      "estimatedImpact": 6,
      "checkScore": 0,
      "checkDetails": "Content ratio: 1.9% (1508 content chars / 79818 HTML bytes)"
    },
    {
      "id": "add_markdown_negotiation",
      "priority": "critical",
      "category": "aiDiscoverability",
      "titleKey": "rec.add_markdown_negotiation.title",
      "descriptionKey": "rec.add_markdown_negotiation.description",
      "howToKey": "rec.add_markdown_negotiation.howto",
      "effort": "significant",
      "estimatedImpact": 6,
      "checkScore": 0,
      "checkDetails": "No Markdown for Agents support detected"
    },
    {
      "id": "add_content_signals",
      "priority": "critical",
      "category": "aiDiscoverability",
      "titleKey": "rec.add_content_signals.title",
      "descriptionKey": "rec.add_content_signals.description",
      "howToKey": "rec.add_content_signals.howto",
      "effort": "quick-win",
      "estimatedImpact": 5,
      "checkScore": 0,
      "checkDetails": "No Content-Signal found (robots.txt or HTTP headers)"
    },
    {
      "id": "remove_inline_styles",
      "priority": "critical",
      "category": "contentEfficiency",
      "titleKey": "rec.remove_inline_styles.title",
      "descriptionKey": "rec.remove_inline_styles.description",
      "howToKey": "rec.remove_inline_styles.howto",
      "effort": "moderate",
      "estimatedImpact": 3,
      "checkScore": 0,
      "checkDetails": "78/640 elements with inline styles (12.2%)"
    },
    {
      "id": "add_semantic_elements",
      "priority": "high",
      "category": "semanticHtml",
      "titleKey": "rec.add_semantic_elements.title",
      "descriptionKey": "rec.add_semantic_elements.description",
      "howToKey": "rec.add_semantic_elements.howto",
      "effort": "moderate",
      "estimatedImpact": 5,
      "checkScore": 25,
      "checkDetails": "14 semantic elements, 169 divs (ratio: 8%)"
    },
    {
      "id": "improve_alt_texts",
      "priority": "high",
      "category": "semanticHtml",
      "titleKey": "rec.improve_alt_texts.title",
      "descriptionKey": "rec.improve_alt_texts.description",
      "howToKey": "rec.improve_alt_texts.howto",
      "effort": "moderate",
      "estimatedImpact": 4,
      "checkScore": 42,
      "checkDetails": "10/24 images with meaningful alt text"
    }
  ],
  "llmsTxtPreview": "# Vast AI\n\n> Rent high-performance cloud GPUs at low cost with Vast.ai. Instantly deploy GPU rentals for AI, machine learning, deep learning, and rendering. Flexible pricing, fast setup, and global availability\n\n## Main\n- [GPU Infrastructure for AI Agents | Vast.ai](https://vast.ai): Rent high-performance cloud GPUs at low cost with Vast.ai. Instantly deploy GPU rentals for AI, machine learning, deep…\n- [Pricing](https://vast.ai/pricing)\n- [GPU Cloud](https://vast.ai/products/gpu-cloud)\n- [Clusters](https://vast.ai/products/clusters)\n- [Serverless](https://vast.ai/products/serverless)\n- [About](https://vast.ai/about)\n- [Developers](https://vast.ai/developers)\n- [CLI](https://vast.ai/developers/cli)\n- [Python SDK](https://vast.ai/developers/sdk)\n- [API](https://vast.ai/developers/api)\n- [Model Library](https://vast.ai/model-library)\n- [Hosting](https://vast.ai/hosting)\n- [Data Centers](https://vast.ai/data-center-application)\n- [Financing](https://vast.ai/hosting/finance)\n\n## Blog\n- [Blog](https://vast.ai/blog)\n\n## Legal\n- [Terms of Service](https://vast.ai/terms)\n- [Privacy Policy](https://vast.ai/privacy)\n\n## Support\n- [Contact Sales](https://vast.ai/contact-sales)\n- [Get in Touch](https://vast.ai/contact)\n\n",
  "llmsTxtExisting": "# Vast.ai\n\n> Vast.ai is a GPU cloud platform offering on-demand access to thousands of GPUs — RTX 4090s, H100s, B200s, and more — from data centers and independent hosts worldwide. No minimum commitment.\n\nSign up and search GPUs at [https://cloud.vast.ai](https://cloud.vast.ai).\n\n## AI Agents (Claude Code, Cursor, Windsurf, etc.)\n\n- [CLI SKILL.md](https://raw.githubusercontent.com/vast-ai/vast-cli/master/vastai/SKILL.md) — commands, auth, quirks, common errors\n- [SDK SKILL.md](https://raw.githubusercontent.com/vast-ai/vast-cli/master/vastai_sdk/SKILL.md) — Python VastAI class, SyncClient, AsyncClient, Serverless\n\nor install the Vast.ai skill:\n\n```\nnpx skills add vast-ai/vast-cli\n```\n\n## Get Started\n\n- [Quickstart](https://docs.vast.ai/documentation/get-started/quickstart.md)\n- [CLI overview & install](https://docs.vast.ai/cli/get-started.md)\n- [Python SDK](https://docs.vast.ai/sdk/python/quickstart.md)\n- [REST API](https://docs.vast.ai/api-reference/introduction.md)\n- [OpenAPI spec](https://raw.githubusercontent.com/vast-ai/vast-cli/master/openapi/yaml/combined_api.yaml)\n\n## Instances\n\n- [Find and rent](https://docs.vast.ai/documentation/instances/choosing/find-and-rent.md)\n- [Instance types](https://docs.vast.ai/documentation/instances/choosing/instance-types.md)\n- [Reserved instances](https://docs.vast.ai/documentation/instances/choosing/reserved-instances.md)\n- [SSH](https://docs.vast.ai/documentation/instances/connect/ssh.md)\n- [Networking & ports](https://docs.vast.ai/documentation/instances/connect/networking.md)\n- [Storage & volumes](https://docs.vast.ai/documentation/instances/storage/volumes.md)\n- [Data movement](https://docs.vast.ai/documentation/instances/storage/data-movement.md)\n- [Virtual machines](https://docs.vast.ai/documentation/instances/virtual-machines.md)\n- [Pricing](https://docs.vast.ai/documentation/instances/pricing.md)\n- [Manage instances](https://docs.vast.ai/documentation/instances/manage-instances.md)\n\n## Serverless\n\n- [Overview](https://docs.vast.ai/documentation/serverless/overview.md)\n- [Getting started](https://docs.vast.ai/documentation/serverless/getting-started-with-serverless.md)\n- [vLLM](https://docs.vast.ai/documentation/serverless/vllm.md)\n- [ComfyUI](https://docs.vast.ai/documentation/serverless/comfy-ui.md)\n- [Worker states](https://docs.vast.ai/documentation/serverless/worker-states.md)\n- [Pricing](https://docs.vast.ai/documentation/serverless/pricing.md)\n- [SDK overview](https://docs.vast.ai/documentation/serverless/SDKoverview.md)\n\n## API Reference\n\n- [Introduction](https://docs.vast.ai/api-reference/introduction.md)\n- [Creating instances](https://docs.vast.ai/api-reference/creating-instances-with-api.md)\n- [Permissions & authorization](https://docs.vast.ai/api-reference/permissions-and-authorization.md)\n- [Rate limits & errors](https://docs.vast.ai/api-reference/rate-limits-and-errors.md)\n\n## Hosting (Suppliers)\n\n- [Hosting overview](https://docs.vast.ai/documentation/host/hosting-overview.md)\n- [Earning](https://docs.vast.ai/documentation/host/earning.md)\n- [Verification](https://docs.vast.ai/documentation/host/understanding-verification.md)\n- [Payment](https://docs.vast.ai/documentation/host/payment.md)\n\n## Reference\n\n- [Billing](https://docs.vast.ai/documentation/reference/billing.md)\n- [API keys](https://docs.vast.ai/documentation/reference/keys.md)\n- [Account settings](https://docs.vast.ai/documentation/reference/account-settings.md)\n- [FAQ](https://docs.vast.ai/documentation/reference/faq/general.md)\n- [Troubleshooting](https://docs.vast.ai/documentation/reference/troubleshooting.md)\n\n## General Information\n\n- [GPU pricing](https://vast.ai/pricing)\n- [Model library](https://vast.ai/model-library)\n- [Hosting / become a supplier](https://vast.ai/hosting)\n- [Case studies](https://vast.ai/case-studies)\n- [Serverless product](https://vast.ai/products/serverless)\n- [GPU Cloud product](https://vast.ai/products/gpu-cloud)\n- [Clusters product](https://vast.ai/products/clusters)\n- [Developers](https://vast.ai/developers)\n- [CLI landing](https://vast.ai/developers/cli)\n- [SDK landing](https://vast.ai/developers/sdk)\n- [About](https://vast.ai/about)\n- [Blog](https://vast.ai/blog)\n\n## Support\n\n- [Live chat support](https://go.crisp.chat/chat/embed/?website_id=734d7b1a-86fc-470d-b60a-f6d4840573ae): 24/7 live chat support for technical issues, billing questions, and platform guidance.\n- [Discord](https://discord.gg/vast): Community support and discussion.\n- [Docs home](https://docs.vast.ai): Main documentation site.\n- [Docs llms index](https://docs.vast.ai/llms.txt): Full docs index for LLM and agent discovery.\n\n## Optional\n\n- [Full resource index](https://vast.ai/llms-full.txt): Expanded catalog of platform pages, GPU pages, documentation, and articles.",
  "emergingProtocols": {
    "oauthDiscovery": {
      "exists": false,
      "url": "https://vast.ai/.well-known/oauth-authorization-server"
    },
    "mcpServerCard": {
      "exists": false,
      "url": "https://vast.ai/.well-known/mcp.json"
    },
    "a2aAgentCard": {
      "exists": false,
      "url": "https://vast.ai/.well-known/agent.json"
    },
    "count": 0
  },
  "snippets": [
    {
      "id": "add_content_signals",
      "title": "Add Content-Signal directives",
      "description": "Content-Signal tells AI agents how they may use your content. The canonical location is robots.txt, but you can also expose it as an HTTP header from any stack.",
      "language": "txt",
      "code": "User-agent: *\nContent-Signal: search=yes, ai-input=yes, ai-train=no",
      "filename": "/robots.txt",
      "stacks": [
        {
          "id": "robots",
          "label": "robots.txt",
          "language": "txt",
          "filename": "/robots.txt",
          "code": "User-agent: *\nContent-Signal: search=yes, ai-input=yes, ai-train=no"
        },
        {
          "id": "nginx",
          "label": "Nginx",
          "language": "nginx",
          "filename": "server block",
          "code": "# Inside your server { } block:\nadd_header Content-Signal \"search=yes, ai-input=yes, ai-train=no\" always;"
        },
        {
          "id": "apache",
          "label": "Apache",
          "language": "apache",
          "filename": ".htaccess",
          "code": "# In .htaccess (or VirtualHost):\nHeader set Content-Signal \"search=yes, ai-input=yes, ai-train=no\""
        },
        {
          "id": "wordpress",
          "label": "WordPress",
          "language": "php",
          "filename": "functions.php",
          "code": "<?php\n// In your theme's functions.php or a small mu-plugin\nadd_action('send_headers', function () {\n    header('Content-Signal: search=yes, ai-input=yes, ai-train=no');\n});\n\n// Optional: also append the directive to the dynamic robots.txt\nadd_filter('robots_txt', function ($output) {\n    return $output . \"\\nContent-Signal: search=yes, ai-input=yes, ai-train=no\\n\";\n}, 10, 1);"
        },
        {
          "id": "nextjs",
          "label": "Next.js",
          "language": "typescript",
          "filename": "middleware.ts",
          "code": "// middleware.ts (Next.js 13+ App Router or Pages Router)\nimport { NextResponse } from 'next/server';\nexport function middleware() {\n  const res = NextResponse.next();\n  res.headers.set(\n    'Content-Signal',\n    'search=yes, ai-input=yes, ai-train=no'\n  );\n  return res;\n}\nexport const config = { matcher: '/:path*' };"
        },
        {
          "id": "cloudflare",
          "label": "Cloudflare Workers",
          "language": "javascript",
          "filename": "worker.js",
          "code": "// Cloudflare Worker that proxies your origin and adds the header\nexport default {\n  async fetch(request, env, ctx) {\n    const res = await fetch(request);\n    const newRes = new Response(res.body, res);\n    newRes.headers.set(\n      'Content-Signal',\n      'search=yes, ai-input=yes, ai-train=no'\n    );\n    return newRes;\n  },\n};"
        },
        {
          "id": "express",
          "label": "Express / Fastify",
          "language": "javascript",
          "filename": "server.js",
          "code": "// Express\napp.use((req, res, next) => {\n  res.setHeader('Content-Signal', 'search=yes, ai-input=yes, ai-train=no');\n  next();\n});\n\n// Fastify\nfastify.addHook('onSend', (request, reply, payload, done) => {\n  reply.header('Content-Signal', 'search=yes, ai-input=yes, ai-train=no');\n  done();\n});"
        }
      ]
    },
    {
      "id": "add_markdown_negotiation",
      "title": "Support Markdown for Agents",
      "description": "Let AI agents request a clean Markdown version of any page via content negotiation, .md alternate URLs, link tags or Link headers.",
      "language": "html",
      "code": "<!-- Mechanism 3: link tag advertising the .md alternate -->\n<link rel=\"alternate\" type=\"text/markdown\" href=\"/page.md\">",
      "filename": "<head>",
      "stacks": [
        {
          "id": "html",
          "label": "HTML <head>",
          "language": "html",
          "filename": "<head>",
          "code": "<!-- Mechanism 3: link tag advertising the .md alternate -->\n<link rel=\"alternate\" type=\"text/markdown\" href=\"/page.md\">"
        },
        {
          "id": "express",
          "label": "Express",
          "language": "javascript",
          "filename": "server.js",
          "code": "// Mechanisms 1 + 4: content negotiation + Link header\napp.get('/page', (req, res) => {\n  res.setHeader('Vary', 'Accept');\n  res.setHeader('Link', '</page.md>; rel=\"alternate\"; type=\"text/markdown\"');\n  if ((req.headers.accept || '').includes('text/markdown')) {\n    res.type('text/markdown; charset=utf-8');\n    return res.send(renderMarkdown('page'));\n  }\n  res.render('page');\n});"
        },
        {
          "id": "fastify",
          "label": "Fastify",
          "language": "javascript",
          "filename": "server.js",
          "code": "// Mechanisms 1 + 4: content negotiation + Link header\nfastify.get('/page', async (req, reply) => {\n  reply.header('Vary', 'Accept');\n  reply.header('Link', '</page.md>; rel=\"alternate\"; type=\"text/markdown\"');\n  if ((req.headers.accept || '').includes('text/markdown')) {\n    return reply.type('text/markdown; charset=utf-8').send(renderMarkdown('page'));\n  }\n  return reply.view('/page.ejs');\n});"
        },
        {
          "id": "nextjs",
          "label": "Next.js",
          "language": "typescript",
          "filename": "app/page/route.ts",
          "code": "// Next.js App Router — Route Handler returning Markdown\nimport { NextRequest } from 'next/server';\nimport { renderMarkdown } from '@/lib/md';\nexport async function GET(req: NextRequest) {\n  const accept = req.headers.get('accept') || '';\n  if (accept.includes('text/markdown')) {\n    return new Response(await renderMarkdown('page'), {\n      headers: {\n        'Content-Type': 'text/markdown; charset=utf-8',\n        'Vary': 'Accept',\n      },\n    });\n  }\n  // Fall through to the page component\n  return new Response(null, { status: 404 });\n}"
        },
        {
          "id": "wordpress",
          "label": "WordPress",
          "language": "php",
          "filename": "functions.php",
          "code": "<?php\n// Mechanism 1: respond to Accept: text/markdown on the same URL\nadd_action('template_redirect', function () {\n    if (!is_singular()) return;\n    $accept = $_SERVER['HTTP_ACCEPT'] ?? '';\n    if (strpos($accept, 'text/markdown') === false) return;\n    header('Content-Type: text/markdown; charset=utf-8');\n    header('Vary: Accept');\n    $post = get_queried_object();\n    echo \"# \" . get_the_title($post) . \"\\n\\n\";\n    echo wp_strip_all_tags(apply_filters('the_content', $post->post_content));\n    exit;\n});"
        },
        {
          "id": "static",
          "label": "Hugo / Jekyll / Astro",
          "language": "txt",
          "filename": "static/page.md",
          "code": "# Mechanism 2: serve .md alongside .html\n# Hugo: place page.md in /static/ — built unchanged\n# Jekyll: drop page.md in /assets/ — copied as-is\n# Astro: src/pages/page.md.ts that exports a GET returning markdown\n\n# Then advertise with mechanism 3 in <head>:\n#   <link rel=\"alternate\" type=\"text/markdown\" href=\"/page.md\">"
        }
      ]
    }
  ]
}

API를 사용하여 프로그래밍 방식으로 가져올 수 있습니다 (곧 출시)