[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-holoscan-install-debian":3,"mdc--k1wtmz-key":31,"related-org-nvidia-holoscan-install-debian":1431,"related-repo-nvidia-holoscan-install-debian":1592},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":20,"repoUrl":21,"updatedAt":22,"license":23,"forks":24,"topics":25,"repo":26,"sourceUrl":29,"mdContent":30},"holoscan-install-debian","install Holoscan SDK on Ubuntu","Install Holoscan SDK natively on Ubuntu via apt. Use for C++ installs on Ubuntu; pair with \u002Fholoscan-install-wheel for Python.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,14,17],{"name":9,"slug":8,"type":13},"tag",{"name":15,"slug":16,"type":13},"Engineering","engineering",{"name":18,"slug":19,"type":13},"SDK","sdk",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-14T05:26:26.438765","Apache-2.0",281,[],{"repoUrl":21,"stars":20,"forks":24,"topics":27,"description":28},[],"AI agent skills published by NVIDIA","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fholoscan-install-debian","---\nname: holoscan-install-debian\nversion: \"1.0.0\"\ndescription: \"Install Holoscan SDK natively on Ubuntu via apt. Use for C++ installs on Ubuntu; pair with \u002Fholoscan-install-wheel for Python.\"\nlicense: Apache-2.0\nmetadata:\n  author: \"Holoscan Team \u003Choloscan-team@nvidia.com>\"\n  github-url: \"https:\u002F\u002Fgithub.com\u002Fnvidia-holoscan\u002Fholoscan-sdk\"\n  tags:\n    - holoscan\n    - install\n    - debian\n    - apt\n    - ubuntu\n---\n\n# Holoscan Debian\u002Fapt Installation\n\n## Purpose\n\nInstall the Holoscan SDK C++ runtime + headers on Ubuntu using NVIDIA's apt repo, selecting the right `holoscan-cuda-*` package for the host's CUDA driver and verifying with the bundled C++ examples.\n\n## Prerequisites\n\n- Ubuntu x86_64 (22.04 \u002F 24.04) or ARM64 (Jetson \u002F IGX) with an NVIDIA GPU and working driver (`nvidia-smi`).\n- `sudo` and network access to `developer.download.nvidia.com` and `docs.nvidia.com`.\n- `cuda-keyring` package (Step 2 installs it if missing).\n\n## Limitations\n\n- No Python bindings from apt — pair with `\u002Fholoscan-install-wheel` if the user needs Python.\n- Ubuntu-only. Other distros must use the container or wheel install.\n- Package variant must match the host CUDA driver (`holoscan-cuda-12` vs `holoscan-cuda-13`); wrong variant → \"CUDA driver version is insufficient\".\n\n## Step 0: Consult the Official Install Instructions\n\nFetch the Debian\u002Fapt section of `https:\u002F\u002Fdocs.nvidia.com\u002Fholoscan\u002Fsdk-user-guide\u002Fsdk_installation.html` before installing. Extract:\n\n- Exact package names (`holoscan-cuda-12`, `holoscan-cuda-13`, `holoscan`)\n- Supported Ubuntu versions\n- The cuda-keyring URL for the right distro\n\nIf the doc disagrees with anything below, the doc wins.\n\nDetermine OS version and CUDA variant if not already known — run in parallel:\n\n```bash\nlsb_release -a 2>\u002Fdev\u002Fnull || cat \u002Fetc\u002Fos-release\nnvidia-smi 2>&1 | head -5\n```\n\n**CUDA variant rule — pick the apt package:**\n\n| nvidia-smi CUDA Version | Package |\n|------------------------|---------|\n| 13.x+ | `holoscan-cuda-13` |\n| 12.x (on IGX) | `holoscan` |\n| 12.x (not on IGX) | `holoscan-cuda-12` |\n| 12.x (nvgpu) | `holoscan-cuda-12` |\n\n## Step 1: Prerequisites Check\n\n```bash\ndpkg -l | grep cuda-keyring\ndpkg -l | grep -E \"holoscan-cuda-(12|13)|^ii  holoscan \"\napt-cache show holoscan-cuda-13 holoscan-cuda-12 2>\u002Fdev\u002Fnull | grep -E \"^(Package|Version)\"\n```\n\nDecision rules based on what Step 1 found:\n\n- Skip the keyring step if `cuda-keyring` is already installed.\n- Skip `apt-get update` if the repo is already configured and the package is visible in `apt-cache show`.\n- **Skip Step 2 entirely** and proceed directly to Step 3 if the correct package variant is already installed (e.g. `holoscan-cuda-12` when targeting cu12).\n\n## Step 2: Install\n\nSkip this step if the package is already installed (detected in Step 1) or if user is on IGX platform.\n\n```bash\n# If cuda-keyring missing (adjust ubuntu2204\u002Fubuntu2404 as needed) and not on IGX platform:\nwget https:\u002F\u002Fdeveloper.download.nvidia.com\u002Fcompute\u002Fcuda\u002Frepos\u002Fubuntu2404\u002Fx86_64\u002Fcuda-keyring_1.1-1_all.deb\nsudo dpkg -i cuda-keyring_1.1-1_all.deb && sudo apt-get update\n\nsudo apt-get install -y holoscan-cuda-12   # or holoscan-cuda-13\n```\n\n## Step 3: Verify\n\nSet the env once for the rest of this step, then run the three C++ checks:\n\n```bash\nHS=\u002Fopt\u002Fnvidia\u002Fholoscan\nexport LD_LIBRARY_PATH=$HS\u002Flib\nexport HOLOSCAN_INPUT_PATH=$HS\u002Fdata\nulimit -s 32768\n\nls $HS\u002Fexamples\u002F{hello_world,tensor_interop,video_replayer}\u002F\n\n# hello_world — expected: \"Hello World!\"\n$HS\u002Fexamples\u002Fhello_world\u002Fcpp\u002Fhello_world\n\n# tensor_interop — expected: tensors doubling each pass, \"Graph execution finished.\"\n# If \"CUDA driver version is insufficient\": swap package variant:\n#   sudo apt-get remove -y holoscan-cuda-13 && sudo apt-get install -y holoscan-cuda-12\n$HS\u002Fexamples\u002Ftensor_interop\u002Fcpp\u002Ftensor_interop\n\n# video_replayer (10 frames, headless) — expected: Vulkan selects NVIDIA GPU, \"Graph execution finished.\"\n# Always run headless: works with or without a display, avoids GUI failure modes over SSH.\nls $HS\u002Fdata\u002Fracerx 2>\u002Fdev\u002Fnull || sudo $HS\u002Fexamples\u002Fdownload_example_data\npython3 -c \"\nc=open('$HS\u002Fexamples\u002Fvideo_replayer\u002Fcpp\u002Fvideo_replayer.yaml').read()\nc=c.replace('count: 0','count: 10').replace('repeat: true','repeat: false').replace('realtime: true','realtime: false')\nc=c.replace('  width: 854','  headless: true\\n  width: 854')\nopen('\u002Ftmp\u002Fvr.yaml','w').write(c)\"\n$HS\u002Fexamples\u002Fvideo_replayer\u002Fcpp\u002Fvideo_replayer --config \u002Ftmp\u002Fvr.yaml\n```\n\n## Step 4: Give the User the Reusable Env Snippet\n\nOnce verified, share this snippet with user and suggest adding it to their shell startup file (e.g., `~\u002F.bashrc`) if they want it to persist across sessions:\n\n```bash\nexport LD_LIBRARY_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Flib:${LD_LIBRARY_PATH}\nexport HOLOSCAN_INPUT_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Fdata\nulimit -s 32768\n```\n\nThen offer next steps:\n- Add Python support: `\u002Fholoscan-install-wheel`\n- Explore examples: `ls \u002Fopt\u002Fnvidia\u002Fholoscan\u002Fexamples\u002F`\n- Walk through a specific example: `\u002Fexplain-example`\n- Start building a custom Holoscan application\n\n## Troubleshooting\n\n- **`python3 -c \"import holoscan\"` fails after apt install.** Expected — the Debian package has been C++ only since v3.0.0. Run `\u002Fholoscan-install-wheel` to add Python bindings.\n- **\"CUDA driver version is insufficient\" when running an example.** Wrong package variant. Re-check `nvidia-smi` CUDA Version and swap variants: `sudo apt-get remove -y holoscan-cuda-13 && sudo apt-get install -y holoscan-cuda-12` (or vice versa).\n- **`E: Unable to locate package holoscan-cuda-12`.** `cuda-keyring` not installed or repo not yet pulled. Run the keyring + `apt-get update` block in Step 2 (adjust `ubuntu2204`\u002F`ubuntu2404` to match the host).\n- **Segmentation fault when launching an example.** `ulimit -s 32768` not set in the current shell. Prepend it to the command (Step 3 pattern).\n- **`error while loading shared libraries: libholoscan_core.so`.** `LD_LIBRARY_PATH` is unset. Use the env snippet from Step 4 — `export LD_LIBRARY_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Flib`.\n- **`video_replayer` can't find data.** Set `HOLOSCAN_INPUT_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Fdata`, or run `sudo \u002Fopt\u002Fnvidia\u002Fholoscan\u002Fexamples\u002Fdownload_example_data` to fetch the `racerx` dataset.\n",{"data":32,"body":43},{"name":4,"version":33,"description":6,"license":23,"metadata":34},"1.0.0",{"author":35,"github-url":36,"tags":37},"Holoscan Team \u003Choloscan-team@nvidia.com>","https:\u002F\u002Fgithub.com\u002Fnvidia-holoscan\u002Fholoscan-sdk",[38,39,40,41,42],"holoscan","install","debian","apt","ubuntu",{"type":44,"children":45},"root",[46,55,62,77,83,139,145,187,193,206,244,249,254,337,346,438,444,574,579,631,637,642,754,760,765,1097,1103,1116,1193,1198,1238,1244,1425],{"type":47,"tag":48,"props":49,"children":51},"element","h1",{"id":50},"holoscan-debianapt-installation",[52],{"type":53,"value":54},"text","Holoscan Debian\u002Fapt Installation",{"type":47,"tag":56,"props":57,"children":59},"h2",{"id":58},"purpose",[60],{"type":53,"value":61},"Purpose",{"type":47,"tag":63,"props":64,"children":65},"p",{},[66,68,75],{"type":53,"value":67},"Install the Holoscan SDK C++ runtime + headers on Ubuntu using NVIDIA's apt repo, selecting the right ",{"type":47,"tag":69,"props":70,"children":72},"code",{"className":71},[],[73],{"type":53,"value":74},"holoscan-cuda-*",{"type":53,"value":76}," package for the host's CUDA driver and verifying with the bundled C++ examples.",{"type":47,"tag":56,"props":78,"children":80},{"id":79},"prerequisites",[81],{"type":53,"value":82},"Prerequisites",{"type":47,"tag":84,"props":85,"children":86},"ul",{},[87,101,128],{"type":47,"tag":88,"props":89,"children":90},"li",{},[91,93,99],{"type":53,"value":92},"Ubuntu x86_64 (22.04 \u002F 24.04) or ARM64 (Jetson \u002F IGX) with an NVIDIA GPU and working driver (",{"type":47,"tag":69,"props":94,"children":96},{"className":95},[],[97],{"type":53,"value":98},"nvidia-smi",{"type":53,"value":100},").",{"type":47,"tag":88,"props":102,"children":103},{},[104,110,112,118,120,126],{"type":47,"tag":69,"props":105,"children":107},{"className":106},[],[108],{"type":53,"value":109},"sudo",{"type":53,"value":111}," and network access to ",{"type":47,"tag":69,"props":113,"children":115},{"className":114},[],[116],{"type":53,"value":117},"developer.download.nvidia.com",{"type":53,"value":119}," and ",{"type":47,"tag":69,"props":121,"children":123},{"className":122},[],[124],{"type":53,"value":125},"docs.nvidia.com",{"type":53,"value":127},".",{"type":47,"tag":88,"props":129,"children":130},{},[131,137],{"type":47,"tag":69,"props":132,"children":134},{"className":133},[],[135],{"type":53,"value":136},"cuda-keyring",{"type":53,"value":138}," package (Step 2 installs it if missing).",{"type":47,"tag":56,"props":140,"children":142},{"id":141},"limitations",[143],{"type":53,"value":144},"Limitations",{"type":47,"tag":84,"props":146,"children":147},{},[148,161,166],{"type":47,"tag":88,"props":149,"children":150},{},[151,153,159],{"type":53,"value":152},"No Python bindings from apt — pair with ",{"type":47,"tag":69,"props":154,"children":156},{"className":155},[],[157],{"type":53,"value":158},"\u002Fholoscan-install-wheel",{"type":53,"value":160}," if the user needs Python.",{"type":47,"tag":88,"props":162,"children":163},{},[164],{"type":53,"value":165},"Ubuntu-only. Other distros must use the container or wheel install.",{"type":47,"tag":88,"props":167,"children":168},{},[169,171,177,179,185],{"type":53,"value":170},"Package variant must match the host CUDA driver (",{"type":47,"tag":69,"props":172,"children":174},{"className":173},[],[175],{"type":53,"value":176},"holoscan-cuda-12",{"type":53,"value":178}," vs ",{"type":47,"tag":69,"props":180,"children":182},{"className":181},[],[183],{"type":53,"value":184},"holoscan-cuda-13",{"type":53,"value":186},"); wrong variant → \"CUDA driver version is insufficient\".",{"type":47,"tag":56,"props":188,"children":190},{"id":189},"step-0-consult-the-official-install-instructions",[191],{"type":53,"value":192},"Step 0: Consult the Official Install Instructions",{"type":47,"tag":63,"props":194,"children":195},{},[196,198,204],{"type":53,"value":197},"Fetch the Debian\u002Fapt section of ",{"type":47,"tag":69,"props":199,"children":201},{"className":200},[],[202],{"type":53,"value":203},"https:\u002F\u002Fdocs.nvidia.com\u002Fholoscan\u002Fsdk-user-guide\u002Fsdk_installation.html",{"type":53,"value":205}," before installing. Extract:",{"type":47,"tag":84,"props":207,"children":208},{},[209,234,239],{"type":47,"tag":88,"props":210,"children":211},{},[212,214,219,221,226,227,232],{"type":53,"value":213},"Exact package names (",{"type":47,"tag":69,"props":215,"children":217},{"className":216},[],[218],{"type":53,"value":176},{"type":53,"value":220},", ",{"type":47,"tag":69,"props":222,"children":224},{"className":223},[],[225],{"type":53,"value":184},{"type":53,"value":220},{"type":47,"tag":69,"props":228,"children":230},{"className":229},[],[231],{"type":53,"value":38},{"type":53,"value":233},")",{"type":47,"tag":88,"props":235,"children":236},{},[237],{"type":53,"value":238},"Supported Ubuntu versions",{"type":47,"tag":88,"props":240,"children":241},{},[242],{"type":53,"value":243},"The cuda-keyring URL for the right distro",{"type":47,"tag":63,"props":245,"children":246},{},[247],{"type":53,"value":248},"If the doc disagrees with anything below, the doc wins.",{"type":47,"tag":63,"props":250,"children":251},{},[252],{"type":53,"value":253},"Determine OS version and CUDA variant if not already known — run in parallel:",{"type":47,"tag":255,"props":256,"children":261},"pre",{"className":257,"code":258,"language":259,"meta":260,"style":260},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","lsb_release -a 2>\u002Fdev\u002Fnull || cat \u002Fetc\u002Fos-release\nnvidia-smi 2>&1 | head -5\n","bash","",[262],{"type":47,"tag":69,"props":263,"children":264},{"__ignoreMap":260},[265,309],{"type":47,"tag":266,"props":267,"children":270},"span",{"class":268,"line":269},"line",1,[271,277,283,289,294,299,304],{"type":47,"tag":266,"props":272,"children":274},{"style":273},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[275],{"type":53,"value":276},"lsb_release",{"type":47,"tag":266,"props":278,"children":280},{"style":279},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[281],{"type":53,"value":282}," -a",{"type":47,"tag":266,"props":284,"children":286},{"style":285},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[287],{"type":53,"value":288}," 2>",{"type":47,"tag":266,"props":290,"children":291},{"style":279},[292],{"type":53,"value":293},"\u002Fdev\u002Fnull",{"type":47,"tag":266,"props":295,"children":296},{"style":285},[297],{"type":53,"value":298}," ||",{"type":47,"tag":266,"props":300,"children":301},{"style":273},[302],{"type":53,"value":303}," cat",{"type":47,"tag":266,"props":305,"children":306},{"style":279},[307],{"type":53,"value":308}," \u002Fetc\u002Fos-release\n",{"type":47,"tag":266,"props":310,"children":312},{"class":268,"line":311},2,[313,317,322,327,332],{"type":47,"tag":266,"props":314,"children":315},{"style":273},[316],{"type":53,"value":98},{"type":47,"tag":266,"props":318,"children":319},{"style":285},[320],{"type":53,"value":321}," 2>&1",{"type":47,"tag":266,"props":323,"children":324},{"style":285},[325],{"type":53,"value":326}," |",{"type":47,"tag":266,"props":328,"children":329},{"style":273},[330],{"type":53,"value":331}," head",{"type":47,"tag":266,"props":333,"children":334},{"style":279},[335],{"type":53,"value":336}," -5\n",{"type":47,"tag":63,"props":338,"children":339},{},[340],{"type":47,"tag":341,"props":342,"children":343},"strong",{},[344],{"type":53,"value":345},"CUDA variant rule — pick the apt package:",{"type":47,"tag":347,"props":348,"children":349},"table",{},[350,369],{"type":47,"tag":351,"props":352,"children":353},"thead",{},[354],{"type":47,"tag":355,"props":356,"children":357},"tr",{},[358,364],{"type":47,"tag":359,"props":360,"children":361},"th",{},[362],{"type":53,"value":363},"nvidia-smi CUDA Version",{"type":47,"tag":359,"props":365,"children":366},{},[367],{"type":53,"value":368},"Package",{"type":47,"tag":370,"props":371,"children":372},"tbody",{},[373,390,406,422],{"type":47,"tag":355,"props":374,"children":375},{},[376,382],{"type":47,"tag":377,"props":378,"children":379},"td",{},[380],{"type":53,"value":381},"13.x+",{"type":47,"tag":377,"props":383,"children":384},{},[385],{"type":47,"tag":69,"props":386,"children":388},{"className":387},[],[389],{"type":53,"value":184},{"type":47,"tag":355,"props":391,"children":392},{},[393,398],{"type":47,"tag":377,"props":394,"children":395},{},[396],{"type":53,"value":397},"12.x (on IGX)",{"type":47,"tag":377,"props":399,"children":400},{},[401],{"type":47,"tag":69,"props":402,"children":404},{"className":403},[],[405],{"type":53,"value":38},{"type":47,"tag":355,"props":407,"children":408},{},[409,414],{"type":47,"tag":377,"props":410,"children":411},{},[412],{"type":53,"value":413},"12.x (not on IGX)",{"type":47,"tag":377,"props":415,"children":416},{},[417],{"type":47,"tag":69,"props":418,"children":420},{"className":419},[],[421],{"type":53,"value":176},{"type":47,"tag":355,"props":423,"children":424},{},[425,430],{"type":47,"tag":377,"props":426,"children":427},{},[428],{"type":53,"value":429},"12.x (nvgpu)",{"type":47,"tag":377,"props":431,"children":432},{},[433],{"type":47,"tag":69,"props":434,"children":436},{"className":435},[],[437],{"type":53,"value":176},{"type":47,"tag":56,"props":439,"children":441},{"id":440},"step-1-prerequisites-check",[442],{"type":53,"value":443},"Step 1: Prerequisites Check",{"type":47,"tag":255,"props":445,"children":447},{"className":257,"code":446,"language":259,"meta":260,"style":260},"dpkg -l | grep cuda-keyring\ndpkg -l | grep -E \"holoscan-cuda-(12|13)|^ii  holoscan \"\napt-cache show holoscan-cuda-13 holoscan-cuda-12 2>\u002Fdev\u002Fnull | grep -E \"^(Package|Version)\"\n",[448],{"type":47,"tag":69,"props":449,"children":450},{"__ignoreMap":260},[451,478,517],{"type":47,"tag":266,"props":452,"children":453},{"class":268,"line":269},[454,459,464,468,473],{"type":47,"tag":266,"props":455,"children":456},{"style":273},[457],{"type":53,"value":458},"dpkg",{"type":47,"tag":266,"props":460,"children":461},{"style":279},[462],{"type":53,"value":463}," -l",{"type":47,"tag":266,"props":465,"children":466},{"style":285},[467],{"type":53,"value":326},{"type":47,"tag":266,"props":469,"children":470},{"style":273},[471],{"type":53,"value":472}," grep",{"type":47,"tag":266,"props":474,"children":475},{"style":279},[476],{"type":53,"value":477}," cuda-keyring\n",{"type":47,"tag":266,"props":479,"children":480},{"class":268,"line":311},[481,485,489,493,497,502,507,512],{"type":47,"tag":266,"props":482,"children":483},{"style":273},[484],{"type":53,"value":458},{"type":47,"tag":266,"props":486,"children":487},{"style":279},[488],{"type":53,"value":463},{"type":47,"tag":266,"props":490,"children":491},{"style":285},[492],{"type":53,"value":326},{"type":47,"tag":266,"props":494,"children":495},{"style":273},[496],{"type":53,"value":472},{"type":47,"tag":266,"props":498,"children":499},{"style":279},[500],{"type":53,"value":501}," -E",{"type":47,"tag":266,"props":503,"children":504},{"style":285},[505],{"type":53,"value":506}," \"",{"type":47,"tag":266,"props":508,"children":509},{"style":279},[510],{"type":53,"value":511},"holoscan-cuda-(12|13)|^ii  holoscan ",{"type":47,"tag":266,"props":513,"children":514},{"style":285},[515],{"type":53,"value":516},"\"\n",{"type":47,"tag":266,"props":518,"children":520},{"class":268,"line":519},3,[521,526,531,536,541,545,549,553,557,561,565,570],{"type":47,"tag":266,"props":522,"children":523},{"style":273},[524],{"type":53,"value":525},"apt-cache",{"type":47,"tag":266,"props":527,"children":528},{"style":279},[529],{"type":53,"value":530}," show",{"type":47,"tag":266,"props":532,"children":533},{"style":279},[534],{"type":53,"value":535}," holoscan-cuda-13",{"type":47,"tag":266,"props":537,"children":538},{"style":279},[539],{"type":53,"value":540}," holoscan-cuda-12",{"type":47,"tag":266,"props":542,"children":543},{"style":285},[544],{"type":53,"value":288},{"type":47,"tag":266,"props":546,"children":547},{"style":279},[548],{"type":53,"value":293},{"type":47,"tag":266,"props":550,"children":551},{"style":285},[552],{"type":53,"value":326},{"type":47,"tag":266,"props":554,"children":555},{"style":273},[556],{"type":53,"value":472},{"type":47,"tag":266,"props":558,"children":559},{"style":279},[560],{"type":53,"value":501},{"type":47,"tag":266,"props":562,"children":563},{"style":285},[564],{"type":53,"value":506},{"type":47,"tag":266,"props":566,"children":567},{"style":279},[568],{"type":53,"value":569},"^(Package|Version)",{"type":47,"tag":266,"props":571,"children":572},{"style":285},[573],{"type":53,"value":516},{"type":47,"tag":63,"props":575,"children":576},{},[577],{"type":53,"value":578},"Decision rules based on what Step 1 found:",{"type":47,"tag":84,"props":580,"children":581},{},[582,594,614],{"type":47,"tag":88,"props":583,"children":584},{},[585,587,592],{"type":53,"value":586},"Skip the keyring step if ",{"type":47,"tag":69,"props":588,"children":590},{"className":589},[],[591],{"type":53,"value":136},{"type":53,"value":593}," is already installed.",{"type":47,"tag":88,"props":595,"children":596},{},[597,599,605,607,613],{"type":53,"value":598},"Skip ",{"type":47,"tag":69,"props":600,"children":602},{"className":601},[],[603],{"type":53,"value":604},"apt-get update",{"type":53,"value":606}," if the repo is already configured and the package is visible in ",{"type":47,"tag":69,"props":608,"children":610},{"className":609},[],[611],{"type":53,"value":612},"apt-cache show",{"type":53,"value":127},{"type":47,"tag":88,"props":615,"children":616},{},[617,622,624,629],{"type":47,"tag":341,"props":618,"children":619},{},[620],{"type":53,"value":621},"Skip Step 2 entirely",{"type":53,"value":623}," and proceed directly to Step 3 if the correct package variant is already installed (e.g. ",{"type":47,"tag":69,"props":625,"children":627},{"className":626},[],[628],{"type":53,"value":176},{"type":53,"value":630}," when targeting cu12).",{"type":47,"tag":56,"props":632,"children":634},{"id":633},"step-2-install",[635],{"type":53,"value":636},"Step 2: Install",{"type":47,"tag":63,"props":638,"children":639},{},[640],{"type":53,"value":641},"Skip this step if the package is already installed (detected in Step 1) or if user is on IGX platform.",{"type":47,"tag":255,"props":643,"children":645},{"className":257,"code":644,"language":259,"meta":260,"style":260},"# If cuda-keyring missing (adjust ubuntu2204\u002Fubuntu2404 as needed) and not on IGX platform:\nwget https:\u002F\u002Fdeveloper.download.nvidia.com\u002Fcompute\u002Fcuda\u002Frepos\u002Fubuntu2404\u002Fx86_64\u002Fcuda-keyring_1.1-1_all.deb\nsudo dpkg -i cuda-keyring_1.1-1_all.deb && sudo apt-get update\n\nsudo apt-get install -y holoscan-cuda-12   # or holoscan-cuda-13\n",[646],{"type":47,"tag":69,"props":647,"children":648},{"__ignoreMap":260},[649,658,671,713,723],{"type":47,"tag":266,"props":650,"children":651},{"class":268,"line":269},[652],{"type":47,"tag":266,"props":653,"children":655},{"style":654},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[656],{"type":53,"value":657},"# If cuda-keyring missing (adjust ubuntu2204\u002Fubuntu2404 as needed) and not on IGX platform:\n",{"type":47,"tag":266,"props":659,"children":660},{"class":268,"line":311},[661,666],{"type":47,"tag":266,"props":662,"children":663},{"style":273},[664],{"type":53,"value":665},"wget",{"type":47,"tag":266,"props":667,"children":668},{"style":279},[669],{"type":53,"value":670}," https:\u002F\u002Fdeveloper.download.nvidia.com\u002Fcompute\u002Fcuda\u002Frepos\u002Fubuntu2404\u002Fx86_64\u002Fcuda-keyring_1.1-1_all.deb\n",{"type":47,"tag":266,"props":672,"children":673},{"class":268,"line":519},[674,678,683,688,693,698,703,708],{"type":47,"tag":266,"props":675,"children":676},{"style":273},[677],{"type":53,"value":109},{"type":47,"tag":266,"props":679,"children":680},{"style":279},[681],{"type":53,"value":682}," dpkg",{"type":47,"tag":266,"props":684,"children":685},{"style":279},[686],{"type":53,"value":687}," -i",{"type":47,"tag":266,"props":689,"children":690},{"style":279},[691],{"type":53,"value":692}," cuda-keyring_1.1-1_all.deb",{"type":47,"tag":266,"props":694,"children":695},{"style":285},[696],{"type":53,"value":697}," &&",{"type":47,"tag":266,"props":699,"children":700},{"style":273},[701],{"type":53,"value":702}," sudo",{"type":47,"tag":266,"props":704,"children":705},{"style":279},[706],{"type":53,"value":707}," apt-get",{"type":47,"tag":266,"props":709,"children":710},{"style":279},[711],{"type":53,"value":712}," update\n",{"type":47,"tag":266,"props":714,"children":716},{"class":268,"line":715},4,[717],{"type":47,"tag":266,"props":718,"children":720},{"emptyLinePlaceholder":719},true,[721],{"type":53,"value":722},"\n",{"type":47,"tag":266,"props":724,"children":726},{"class":268,"line":725},5,[727,731,735,740,745,749],{"type":47,"tag":266,"props":728,"children":729},{"style":273},[730],{"type":53,"value":109},{"type":47,"tag":266,"props":732,"children":733},{"style":279},[734],{"type":53,"value":707},{"type":47,"tag":266,"props":736,"children":737},{"style":279},[738],{"type":53,"value":739}," install",{"type":47,"tag":266,"props":741,"children":742},{"style":279},[743],{"type":53,"value":744}," -y",{"type":47,"tag":266,"props":746,"children":747},{"style":279},[748],{"type":53,"value":540},{"type":47,"tag":266,"props":750,"children":751},{"style":654},[752],{"type":53,"value":753},"   # or holoscan-cuda-13\n",{"type":47,"tag":56,"props":755,"children":757},{"id":756},"step-3-verify",[758],{"type":53,"value":759},"Step 3: Verify",{"type":47,"tag":63,"props":761,"children":762},{},[763],{"type":53,"value":764},"Set the env once for the rest of this step, then run the three C++ checks:",{"type":47,"tag":255,"props":766,"children":768},{"className":257,"code":767,"language":259,"meta":260,"style":260},"HS=\u002Fopt\u002Fnvidia\u002Fholoscan\nexport LD_LIBRARY_PATH=$HS\u002Flib\nexport HOLOSCAN_INPUT_PATH=$HS\u002Fdata\nulimit -s 32768\n\nls $HS\u002Fexamples\u002F{hello_world,tensor_interop,video_replayer}\u002F\n\n# hello_world — expected: \"Hello World!\"\n$HS\u002Fexamples\u002Fhello_world\u002Fcpp\u002Fhello_world\n\n# tensor_interop — expected: tensors doubling each pass, \"Graph execution finished.\"\n# If \"CUDA driver version is insufficient\": swap package variant:\n#   sudo apt-get remove -y holoscan-cuda-13 && sudo apt-get install -y holoscan-cuda-12\n$HS\u002Fexamples\u002Ftensor_interop\u002Fcpp\u002Ftensor_interop\n\n# video_replayer (10 frames, headless) — expected: Vulkan selects NVIDIA GPU, \"Graph execution finished.\"\n# Always run headless: works with or without a display, avoids GUI failure modes over SSH.\nls $HS\u002Fdata\u002Fracerx 2>\u002Fdev\u002Fnull || sudo $HS\u002Fexamples\u002Fdownload_example_data\npython3 -c \"\nc=open('$HS\u002Fexamples\u002Fvideo_replayer\u002Fcpp\u002Fvideo_replayer.yaml').read()\nc=c.replace('count: 0','count: 10').replace('repeat: true','repeat: false').replace('realtime: true','realtime: false')\nc=c.replace('  width: 854','  headless: true\\n  width: 854')\nopen('\u002Ftmp\u002Fvr.yaml','w').write(c)\"\n$HS\u002Fexamples\u002Fvideo_replayer\u002Fcpp\u002Fvideo_replayer --config \u002Ftmp\u002Fvr.yaml\n",[769],{"type":47,"tag":69,"props":770,"children":771},{"__ignoreMap":260},[772,791,814,835,855,862,881,889,898,907,915,924,933,942,951,959,968,977,1019,1038,1057,1066,1075,1088],{"type":47,"tag":266,"props":773,"children":774},{"class":268,"line":269},[775,781,786],{"type":47,"tag":266,"props":776,"children":778},{"style":777},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[779],{"type":53,"value":780},"HS",{"type":47,"tag":266,"props":782,"children":783},{"style":285},[784],{"type":53,"value":785},"=",{"type":47,"tag":266,"props":787,"children":788},{"style":279},[789],{"type":53,"value":790},"\u002Fopt\u002Fnvidia\u002Fholoscan\n",{"type":47,"tag":266,"props":792,"children":793},{"class":268,"line":311},[794,800,805,809],{"type":47,"tag":266,"props":795,"children":797},{"style":796},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[798],{"type":53,"value":799},"export",{"type":47,"tag":266,"props":801,"children":802},{"style":777},[803],{"type":53,"value":804}," LD_LIBRARY_PATH",{"type":47,"tag":266,"props":806,"children":807},{"style":285},[808],{"type":53,"value":785},{"type":47,"tag":266,"props":810,"children":811},{"style":777},[812],{"type":53,"value":813},"$HS\u002Flib\n",{"type":47,"tag":266,"props":815,"children":816},{"class":268,"line":519},[817,821,826,830],{"type":47,"tag":266,"props":818,"children":819},{"style":796},[820],{"type":53,"value":799},{"type":47,"tag":266,"props":822,"children":823},{"style":777},[824],{"type":53,"value":825}," HOLOSCAN_INPUT_PATH",{"type":47,"tag":266,"props":827,"children":828},{"style":285},[829],{"type":53,"value":785},{"type":47,"tag":266,"props":831,"children":832},{"style":777},[833],{"type":53,"value":834},"$HS\u002Fdata\n",{"type":47,"tag":266,"props":836,"children":837},{"class":268,"line":715},[838,844,849],{"type":47,"tag":266,"props":839,"children":841},{"style":840},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[842],{"type":53,"value":843},"ulimit",{"type":47,"tag":266,"props":845,"children":846},{"style":279},[847],{"type":53,"value":848}," -s",{"type":47,"tag":266,"props":850,"children":852},{"style":851},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[853],{"type":53,"value":854}," 32768\n",{"type":47,"tag":266,"props":856,"children":857},{"class":268,"line":725},[858],{"type":47,"tag":266,"props":859,"children":860},{"emptyLinePlaceholder":719},[861],{"type":53,"value":722},{"type":47,"tag":266,"props":863,"children":865},{"class":268,"line":864},6,[866,871,876],{"type":47,"tag":266,"props":867,"children":868},{"style":273},[869],{"type":53,"value":870},"ls",{"type":47,"tag":266,"props":872,"children":873},{"style":777},[874],{"type":53,"value":875}," $HS",{"type":47,"tag":266,"props":877,"children":878},{"style":279},[879],{"type":53,"value":880},"\u002Fexamples\u002F{hello_world,tensor_interop,video_replayer}\u002F\n",{"type":47,"tag":266,"props":882,"children":884},{"class":268,"line":883},7,[885],{"type":47,"tag":266,"props":886,"children":887},{"emptyLinePlaceholder":719},[888],{"type":53,"value":722},{"type":47,"tag":266,"props":890,"children":892},{"class":268,"line":891},8,[893],{"type":47,"tag":266,"props":894,"children":895},{"style":654},[896],{"type":53,"value":897},"# hello_world — expected: \"Hello World!\"\n",{"type":47,"tag":266,"props":899,"children":901},{"class":268,"line":900},9,[902],{"type":47,"tag":266,"props":903,"children":904},{"style":777},[905],{"type":53,"value":906},"$HS\u002Fexamples\u002Fhello_world\u002Fcpp\u002Fhello_world\n",{"type":47,"tag":266,"props":908,"children":910},{"class":268,"line":909},10,[911],{"type":47,"tag":266,"props":912,"children":913},{"emptyLinePlaceholder":719},[914],{"type":53,"value":722},{"type":47,"tag":266,"props":916,"children":918},{"class":268,"line":917},11,[919],{"type":47,"tag":266,"props":920,"children":921},{"style":654},[922],{"type":53,"value":923},"# tensor_interop — expected: tensors doubling each pass, \"Graph execution finished.\"\n",{"type":47,"tag":266,"props":925,"children":927},{"class":268,"line":926},12,[928],{"type":47,"tag":266,"props":929,"children":930},{"style":654},[931],{"type":53,"value":932},"# If \"CUDA driver version is insufficient\": swap package variant:\n",{"type":47,"tag":266,"props":934,"children":936},{"class":268,"line":935},13,[937],{"type":47,"tag":266,"props":938,"children":939},{"style":654},[940],{"type":53,"value":941},"#   sudo apt-get remove -y holoscan-cuda-13 && sudo apt-get install -y holoscan-cuda-12\n",{"type":47,"tag":266,"props":943,"children":945},{"class":268,"line":944},14,[946],{"type":47,"tag":266,"props":947,"children":948},{"style":777},[949],{"type":53,"value":950},"$HS\u002Fexamples\u002Ftensor_interop\u002Fcpp\u002Ftensor_interop\n",{"type":47,"tag":266,"props":952,"children":954},{"class":268,"line":953},15,[955],{"type":47,"tag":266,"props":956,"children":957},{"emptyLinePlaceholder":719},[958],{"type":53,"value":722},{"type":47,"tag":266,"props":960,"children":962},{"class":268,"line":961},16,[963],{"type":47,"tag":266,"props":964,"children":965},{"style":654},[966],{"type":53,"value":967},"# video_replayer (10 frames, headless) — expected: Vulkan selects NVIDIA GPU, \"Graph execution finished.\"\n",{"type":47,"tag":266,"props":969,"children":971},{"class":268,"line":970},17,[972],{"type":47,"tag":266,"props":973,"children":974},{"style":654},[975],{"type":53,"value":976},"# Always run headless: works with or without a display, avoids GUI failure modes over SSH.\n",{"type":47,"tag":266,"props":978,"children":980},{"class":268,"line":979},18,[981,985,989,994,998,1002,1006,1010,1014],{"type":47,"tag":266,"props":982,"children":983},{"style":273},[984],{"type":53,"value":870},{"type":47,"tag":266,"props":986,"children":987},{"style":777},[988],{"type":53,"value":875},{"type":47,"tag":266,"props":990,"children":991},{"style":279},[992],{"type":53,"value":993},"\u002Fdata\u002Fracerx",{"type":47,"tag":266,"props":995,"children":996},{"style":285},[997],{"type":53,"value":288},{"type":47,"tag":266,"props":999,"children":1000},{"style":279},[1001],{"type":53,"value":293},{"type":47,"tag":266,"props":1003,"children":1004},{"style":285},[1005],{"type":53,"value":298},{"type":47,"tag":266,"props":1007,"children":1008},{"style":273},[1009],{"type":53,"value":702},{"type":47,"tag":266,"props":1011,"children":1012},{"style":777},[1013],{"type":53,"value":875},{"type":47,"tag":266,"props":1015,"children":1016},{"style":279},[1017],{"type":53,"value":1018},"\u002Fexamples\u002Fdownload_example_data\n",{"type":47,"tag":266,"props":1020,"children":1022},{"class":268,"line":1021},19,[1023,1028,1033],{"type":47,"tag":266,"props":1024,"children":1025},{"style":273},[1026],{"type":53,"value":1027},"python3",{"type":47,"tag":266,"props":1029,"children":1030},{"style":279},[1031],{"type":53,"value":1032}," -c",{"type":47,"tag":266,"props":1034,"children":1035},{"style":285},[1036],{"type":53,"value":1037}," \"\n",{"type":47,"tag":266,"props":1039,"children":1041},{"class":268,"line":1040},20,[1042,1047,1052],{"type":47,"tag":266,"props":1043,"children":1044},{"style":279},[1045],{"type":53,"value":1046},"c=open('",{"type":47,"tag":266,"props":1048,"children":1049},{"style":777},[1050],{"type":53,"value":1051},"$HS",{"type":47,"tag":266,"props":1053,"children":1054},{"style":279},[1055],{"type":53,"value":1056},"\u002Fexamples\u002Fvideo_replayer\u002Fcpp\u002Fvideo_replayer.yaml').read()\n",{"type":47,"tag":266,"props":1058,"children":1060},{"class":268,"line":1059},21,[1061],{"type":47,"tag":266,"props":1062,"children":1063},{"style":279},[1064],{"type":53,"value":1065},"c=c.replace('count: 0','count: 10').replace('repeat: true','repeat: false').replace('realtime: true','realtime: false')\n",{"type":47,"tag":266,"props":1067,"children":1069},{"class":268,"line":1068},22,[1070],{"type":47,"tag":266,"props":1071,"children":1072},{"style":279},[1073],{"type":53,"value":1074},"c=c.replace('  width: 854','  headless: true\\n  width: 854')\n",{"type":47,"tag":266,"props":1076,"children":1078},{"class":268,"line":1077},23,[1079,1084],{"type":47,"tag":266,"props":1080,"children":1081},{"style":279},[1082],{"type":53,"value":1083},"open('\u002Ftmp\u002Fvr.yaml','w').write(c)",{"type":47,"tag":266,"props":1085,"children":1086},{"style":285},[1087],{"type":53,"value":516},{"type":47,"tag":266,"props":1089,"children":1091},{"class":268,"line":1090},24,[1092],{"type":47,"tag":266,"props":1093,"children":1094},{"style":777},[1095],{"type":53,"value":1096},"$HS\u002Fexamples\u002Fvideo_replayer\u002Fcpp\u002Fvideo_replayer --config \u002Ftmp\u002Fvr.yaml\n",{"type":47,"tag":56,"props":1098,"children":1100},{"id":1099},"step-4-give-the-user-the-reusable-env-snippet",[1101],{"type":53,"value":1102},"Step 4: Give the User the Reusable Env Snippet",{"type":47,"tag":63,"props":1104,"children":1105},{},[1106,1108,1114],{"type":53,"value":1107},"Once verified, share this snippet with user and suggest adding it to their shell startup file (e.g., ",{"type":47,"tag":69,"props":1109,"children":1111},{"className":1110},[],[1112],{"type":53,"value":1113},"~\u002F.bashrc",{"type":53,"value":1115},") if they want it to persist across sessions:",{"type":47,"tag":255,"props":1117,"children":1119},{"className":257,"code":1118,"language":259,"meta":260,"style":260},"export LD_LIBRARY_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Flib:${LD_LIBRARY_PATH}\nexport HOLOSCAN_INPUT_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Fdata\nulimit -s 32768\n",[1120],{"type":47,"tag":69,"props":1121,"children":1122},{"__ignoreMap":260},[1123,1158,1178],{"type":47,"tag":266,"props":1124,"children":1125},{"class":268,"line":269},[1126,1130,1134,1138,1143,1148,1153],{"type":47,"tag":266,"props":1127,"children":1128},{"style":796},[1129],{"type":53,"value":799},{"type":47,"tag":266,"props":1131,"children":1132},{"style":777},[1133],{"type":53,"value":804},{"type":47,"tag":266,"props":1135,"children":1136},{"style":285},[1137],{"type":53,"value":785},{"type":47,"tag":266,"props":1139,"children":1140},{"style":777},[1141],{"type":53,"value":1142},"\u002Fopt\u002Fnvidia\u002Fholoscan\u002Flib:",{"type":47,"tag":266,"props":1144,"children":1145},{"style":285},[1146],{"type":53,"value":1147},"${",{"type":47,"tag":266,"props":1149,"children":1150},{"style":777},[1151],{"type":53,"value":1152},"LD_LIBRARY_PATH",{"type":47,"tag":266,"props":1154,"children":1155},{"style":285},[1156],{"type":53,"value":1157},"}\n",{"type":47,"tag":266,"props":1159,"children":1160},{"class":268,"line":311},[1161,1165,1169,1173],{"type":47,"tag":266,"props":1162,"children":1163},{"style":796},[1164],{"type":53,"value":799},{"type":47,"tag":266,"props":1166,"children":1167},{"style":777},[1168],{"type":53,"value":825},{"type":47,"tag":266,"props":1170,"children":1171},{"style":285},[1172],{"type":53,"value":785},{"type":47,"tag":266,"props":1174,"children":1175},{"style":777},[1176],{"type":53,"value":1177},"\u002Fopt\u002Fnvidia\u002Fholoscan\u002Fdata\n",{"type":47,"tag":266,"props":1179,"children":1180},{"class":268,"line":519},[1181,1185,1189],{"type":47,"tag":266,"props":1182,"children":1183},{"style":840},[1184],{"type":53,"value":843},{"type":47,"tag":266,"props":1186,"children":1187},{"style":279},[1188],{"type":53,"value":848},{"type":47,"tag":266,"props":1190,"children":1191},{"style":851},[1192],{"type":53,"value":854},{"type":47,"tag":63,"props":1194,"children":1195},{},[1196],{"type":53,"value":1197},"Then offer next steps:",{"type":47,"tag":84,"props":1199,"children":1200},{},[1201,1211,1222,1233],{"type":47,"tag":88,"props":1202,"children":1203},{},[1204,1206],{"type":53,"value":1205},"Add Python support: ",{"type":47,"tag":69,"props":1207,"children":1209},{"className":1208},[],[1210],{"type":53,"value":158},{"type":47,"tag":88,"props":1212,"children":1213},{},[1214,1216],{"type":53,"value":1215},"Explore examples: ",{"type":47,"tag":69,"props":1217,"children":1219},{"className":1218},[],[1220],{"type":53,"value":1221},"ls \u002Fopt\u002Fnvidia\u002Fholoscan\u002Fexamples\u002F",{"type":47,"tag":88,"props":1223,"children":1224},{},[1225,1227],{"type":53,"value":1226},"Walk through a specific example: ",{"type":47,"tag":69,"props":1228,"children":1230},{"className":1229},[],[1231],{"type":53,"value":1232},"\u002Fexplain-example",{"type":47,"tag":88,"props":1234,"children":1235},{},[1236],{"type":53,"value":1237},"Start building a custom Holoscan application",{"type":47,"tag":56,"props":1239,"children":1241},{"id":1240},"troubleshooting",[1242],{"type":53,"value":1243},"Troubleshooting",{"type":47,"tag":84,"props":1245,"children":1246},{},[1247,1270,1295,1340,1357,1385],{"type":47,"tag":88,"props":1248,"children":1249},{},[1250,1261,1263,1268],{"type":47,"tag":341,"props":1251,"children":1252},{},[1253,1259],{"type":47,"tag":69,"props":1254,"children":1256},{"className":1255},[],[1257],{"type":53,"value":1258},"python3 -c \"import holoscan\"",{"type":53,"value":1260}," fails after apt install.",{"type":53,"value":1262}," Expected — the Debian package has been C++ only since v3.0.0. Run ",{"type":47,"tag":69,"props":1264,"children":1266},{"className":1265},[],[1267],{"type":53,"value":158},{"type":53,"value":1269}," to add Python bindings.",{"type":47,"tag":88,"props":1271,"children":1272},{},[1273,1278,1280,1285,1287,1293],{"type":47,"tag":341,"props":1274,"children":1275},{},[1276],{"type":53,"value":1277},"\"CUDA driver version is insufficient\" when running an example.",{"type":53,"value":1279}," Wrong package variant. Re-check ",{"type":47,"tag":69,"props":1281,"children":1283},{"className":1282},[],[1284],{"type":53,"value":98},{"type":53,"value":1286}," CUDA Version and swap variants: ",{"type":47,"tag":69,"props":1288,"children":1290},{"className":1289},[],[1291],{"type":53,"value":1292},"sudo apt-get remove -y holoscan-cuda-13 && sudo apt-get install -y holoscan-cuda-12",{"type":53,"value":1294}," (or vice versa).",{"type":47,"tag":88,"props":1296,"children":1297},{},[1298,1308,1310,1315,1317,1322,1324,1330,1332,1338],{"type":47,"tag":341,"props":1299,"children":1300},{},[1301,1307],{"type":47,"tag":69,"props":1302,"children":1304},{"className":1303},[],[1305],{"type":53,"value":1306},"E: Unable to locate package holoscan-cuda-12",{"type":53,"value":127},{"type":53,"value":1309}," ",{"type":47,"tag":69,"props":1311,"children":1313},{"className":1312},[],[1314],{"type":53,"value":136},{"type":53,"value":1316}," not installed or repo not yet pulled. Run the keyring + ",{"type":47,"tag":69,"props":1318,"children":1320},{"className":1319},[],[1321],{"type":53,"value":604},{"type":53,"value":1323}," block in Step 2 (adjust ",{"type":47,"tag":69,"props":1325,"children":1327},{"className":1326},[],[1328],{"type":53,"value":1329},"ubuntu2204",{"type":53,"value":1331},"\u002F",{"type":47,"tag":69,"props":1333,"children":1335},{"className":1334},[],[1336],{"type":53,"value":1337},"ubuntu2404",{"type":53,"value":1339}," to match the host).",{"type":47,"tag":88,"props":1341,"children":1342},{},[1343,1348,1349,1355],{"type":47,"tag":341,"props":1344,"children":1345},{},[1346],{"type":53,"value":1347},"Segmentation fault when launching an example.",{"type":53,"value":1309},{"type":47,"tag":69,"props":1350,"children":1352},{"className":1351},[],[1353],{"type":53,"value":1354},"ulimit -s 32768",{"type":53,"value":1356}," not set in the current shell. Prepend it to the command (Step 3 pattern).",{"type":47,"tag":88,"props":1358,"children":1359},{},[1360,1370,1371,1376,1378,1384],{"type":47,"tag":341,"props":1361,"children":1362},{},[1363,1369],{"type":47,"tag":69,"props":1364,"children":1366},{"className":1365},[],[1367],{"type":53,"value":1368},"error while loading shared libraries: libholoscan_core.so",{"type":53,"value":127},{"type":53,"value":1309},{"type":47,"tag":69,"props":1372,"children":1374},{"className":1373},[],[1375],{"type":53,"value":1152},{"type":53,"value":1377}," is unset. Use the env snippet from Step 4 — ",{"type":47,"tag":69,"props":1379,"children":1381},{"className":1380},[],[1382],{"type":53,"value":1383},"export LD_LIBRARY_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Flib",{"type":53,"value":127},{"type":47,"tag":88,"props":1386,"children":1387},{},[1388,1399,1401,1407,1409,1415,1417,1423],{"type":47,"tag":341,"props":1389,"children":1390},{},[1391,1397],{"type":47,"tag":69,"props":1392,"children":1394},{"className":1393},[],[1395],{"type":53,"value":1396},"video_replayer",{"type":53,"value":1398}," can't find data.",{"type":53,"value":1400}," Set ",{"type":47,"tag":69,"props":1402,"children":1404},{"className":1403},[],[1405],{"type":53,"value":1406},"HOLOSCAN_INPUT_PATH=\u002Fopt\u002Fnvidia\u002Fholoscan\u002Fdata",{"type":53,"value":1408},", or run ",{"type":47,"tag":69,"props":1410,"children":1412},{"className":1411},[],[1413],{"type":53,"value":1414},"sudo \u002Fopt\u002Fnvidia\u002Fholoscan\u002Fexamples\u002Fdownload_example_data",{"type":53,"value":1416}," to fetch the ",{"type":47,"tag":69,"props":1418,"children":1420},{"className":1419},[],[1421],{"type":53,"value":1422},"racerx",{"type":53,"value":1424}," dataset.",{"type":47,"tag":1426,"props":1427,"children":1428},"style",{},[1429],{"type":53,"value":1430},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1432,"total":1591},[1433,1451,1469,1480,1492,1506,1519,1533,1546,1557,1571,1580],{"slug":1434,"name":1434,"fn":1435,"description":1436,"org":1437,"tags":1438,"stars":1448,"repoUrl":1449,"updatedAt":1450},"nemoclaw-user-guide","retrieve NemoClaw documentation and configuration","Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1439,1442,1445],{"name":1440,"slug":1441,"type":13},"Documentation","documentation",{"name":1443,"slug":1444,"type":13},"MCP","mcp",{"name":1446,"slug":1447,"type":13},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1452,"name":1452,"fn":1453,"description":1454,"org":1455,"tags":1456,"stars":1466,"repoUrl":1467,"updatedAt":1468},"mcore-build-and-dependency","manage Megatron-LM development environments","Container-based dev environment setup and dependency management for Megatron-LM. Covers acquiring and launching the CI container, uv package management, and updating uv.lock.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1457,1460,1463],{"name":1458,"slug":1459,"type":13},"Containers","containers",{"name":1461,"slug":1462,"type":13},"Deployment","deployment",{"name":1464,"slug":1465,"type":13},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1470,"name":1470,"fn":1471,"description":1472,"org":1473,"tags":1474,"stars":1466,"repoUrl":1467,"updatedAt":1479},"mcore-bump-base-image","update NVIDIA PyTorch base images","Bump the NVIDIA PyTorch base image (`nvcr.io\u002Fnvidia\u002Fpytorch:YY.MM-py3`) used by Megatron-LM CI. Covers the two pin sites (GitHub CI in `docker\u002F.ngc_version.dev` and GitLab CI in `.gitlab\u002Fstages\u002F01.build.yml`), the post-bump CI loop (re-run functional tests, refresh golden values, mark broken tests), and the gotchas that bit PRs",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1475,1478],{"name":1476,"slug":1477,"type":13},"CI\u002FCD","ci-cd",{"name":1461,"slug":1462,"type":13},"2026-07-14T05:25:59.97109",{"slug":1481,"name":1481,"fn":1482,"description":1483,"org":1484,"tags":1485,"stars":1466,"repoUrl":1467,"updatedAt":1491},"mcore-cicd","manage CI\u002FCD pipelines for Megatron-LM","CI\u002FCD reference for Megatron-LM. Covers CI pipeline structure, PR scope labels, triggering internal GitLab CI (which force-pushes the current branch to a pull-request\u002FBRANCH ref — always dry-run and verify the destination first; never run against shared or protected branches), and CI failure investigation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1486,1487,1488],{"name":1476,"slug":1477,"type":13},{"name":1461,"slug":1462,"type":13},{"name":1489,"slug":1490,"type":13},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1493,"name":1493,"fn":1494,"description":1495,"org":1496,"tags":1497,"stars":1466,"repoUrl":1467,"updatedAt":1505},"mcore-create-issue","investigate CI failures and create issues","Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1498,1501,1502],{"name":1499,"slug":1500,"type":13},"Debugging","debugging",{"name":1489,"slug":1490,"type":13},{"name":1503,"slug":1504,"type":13},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1507,"name":1507,"fn":1508,"description":1509,"org":1510,"tags":1511,"stars":1466,"repoUrl":1467,"updatedAt":1518},"mcore-linting-and-formatting","lint and format Megatron-LM code","Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1512,1515],{"name":1513,"slug":1514,"type":13},"Best Practices","best-practices",{"name":1516,"slug":1517,"type":13},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1520,"name":1520,"fn":1521,"description":1522,"org":1523,"tags":1524,"stars":1466,"repoUrl":1467,"updatedAt":1532},"mcore-migrate-gpt-to-hybrid","migrate Megatron-LM models to HybridModel","Migration guide for moving Megatron Core GPTModel checkpoints, model providers, training commands, and layer mappings to HybridModel.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1525,1528,1531],{"name":1526,"slug":1527,"type":13},"Machine Learning","machine-learning",{"name":1529,"slug":1530,"type":13},"Migration","migration",{"name":9,"slug":8,"type":13},"2026-07-17T06:07:11.777011",{"slug":1534,"name":1534,"fn":1535,"description":1536,"org":1537,"tags":1538,"stars":1466,"repoUrl":1467,"updatedAt":1545},"mcore-onboard-gb200-1node-tests","onboard functional tests for GB200","Onboard 1-node GitHub MR functional tests for GB200 from existing mr-scoped 2-node tests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1539,1542],{"name":1540,"slug":1541,"type":13},"QA","qa",{"name":1543,"slug":1544,"type":13},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":1547,"name":1547,"fn":1548,"description":1549,"org":1550,"tags":1551,"stars":1466,"repoUrl":1467,"updatedAt":1556},"mcore-run-on-slurm","launch distributed training jobs on SLURM","How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1552,1553],{"name":1461,"slug":1462,"type":13},{"name":1554,"slug":1555,"type":13},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":1558,"name":1558,"fn":1559,"description":1560,"org":1561,"tags":1562,"stars":1466,"repoUrl":1467,"updatedAt":1570},"mcore-split-pr","split pull requests to reduce review load","Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1563,1566,1567],{"name":1564,"slug":1565,"type":13},"Code Review","code-review",{"name":1489,"slug":1490,"type":13},{"name":1568,"slug":1569,"type":13},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1572,"name":1572,"fn":1573,"description":1574,"org":1575,"tags":1576,"stars":1466,"repoUrl":1467,"updatedAt":1579},"mcore-testing","run and manage Megatron-LM tests","Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1577,1578],{"name":1540,"slug":1541,"type":13},{"name":1543,"slug":1544,"type":13},"2026-07-14T05:25:54.928983",{"slug":1581,"name":1581,"fn":1582,"description":1583,"org":1584,"tags":1585,"stars":1466,"repoUrl":1467,"updatedAt":1590},"nightly-sync","manage nightly main-to-dev sync workflows","Domain knowledge for the nightly main-to-dev sync workflow. Covers merge strategy, CI architecture, failure investigation, and known issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1586,1589],{"name":1587,"slug":1588,"type":13},"Automation","automation",{"name":1476,"slug":1477,"type":13},"2026-07-30T05:29:03.275638",496,{"items":1593,"total":1689},[1594,1611,1621,1635,1645,1660,1675],{"slug":1595,"name":1595,"fn":1596,"description":1597,"org":1598,"tags":1599,"stars":20,"repoUrl":21,"updatedAt":1610},"accelerated-computing-cudf","accelerate data processing with cuDF","Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV\u002FParquet I\u002FO, nullable semantics, and multi-GPU DataFrame workloads.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1600,1603,1606,1607],{"name":1601,"slug":1602,"type":13},"Data Analysis","data-analysis",{"name":1604,"slug":1605,"type":13},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":13},{"name":1608,"slug":1609,"type":13},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":1612,"name":1612,"fn":1613,"description":1614,"org":1615,"tags":1616,"stars":20,"repoUrl":21,"updatedAt":1620},"aiq-deploy","deploy and manage NVIDIA AI-Q infrastructure","Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1617,1618,1619],{"name":1461,"slug":1462,"type":13},{"name":1554,"slug":1555,"type":13},{"name":9,"slug":8,"type":13},"2026-07-14T05:29:06.667109",{"slug":1622,"name":1622,"fn":1623,"description":1624,"org":1625,"tags":1626,"stars":20,"repoUrl":21,"updatedAt":1634},"aiq-research","conduct deep research with AI-Q","Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1627,1630,1631],{"name":1628,"slug":1629,"type":13},"Agents","agents",{"name":9,"slug":8,"type":13},{"name":1632,"slug":1633,"type":13},"Research","research","2026-07-14T05:28:06.816956",{"slug":1636,"name":1636,"fn":1637,"description":1638,"org":1639,"tags":1640,"stars":20,"repoUrl":21,"updatedAt":1644},"amc-run-sample-calibration","run AMC sample dataset calibration","Run end-to-end calibration on the shipped sample dataset (sdg_08_2_sample_data_010926.zip) against a running AMC microservice. Use when user says 'test sample dataset', 'run sample calibration', 'verify AMC install', or 'launch and test'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1641,1642,1643],{"name":1601,"slug":1602,"type":13},{"name":9,"slug":8,"type":13},{"name":1543,"slug":1544,"type":13},"2026-07-17T05:29:03.913266",{"slug":1646,"name":1646,"fn":1647,"description":1648,"org":1649,"tags":1650,"stars":20,"repoUrl":21,"updatedAt":1659},"amc-run-video-calibration","calibrate video datasets with AutoMagicCalib","Calibrate a new dataset from pre-recorded video files via the AutoMagicCalib REST API. Use when user has local MP4s and says 'calibrate my videos', 'run AMC on these videos', or similar. For RTSP\u002Flive streams, use amc-run-rtsp-calibration instead.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1651,1652,1655,1656],{"name":1587,"slug":1588,"type":13},{"name":1653,"slug":1654,"type":13},"Imaging","imaging",{"name":9,"slug":8,"type":13},{"name":1657,"slug":1658,"type":13},"Video","video","2026-07-17T05:28:53.905004",{"slug":1661,"name":1661,"fn":1662,"description":1663,"org":1664,"tags":1665,"stars":20,"repoUrl":21,"updatedAt":1674},"amc-setup-calibration-stack","deploy AutoMagicCalib microservice with Docker","Launch AutoMagicCalib microservice and web UI from NGC release images via Docker Compose. Use when user says 'deploy auto calibration', 'launch auto calibration', 'launch AMC', 'start MS+UI', or 'set up auto-magic-calib'. Requires NGC API key.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1666,1667,1670,1671],{"name":1461,"slug":1462,"type":13},{"name":1668,"slug":1669,"type":13},"Docker","docker",{"name":9,"slug":8,"type":13},{"name":1672,"slug":1673,"type":13},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":1676,"name":1676,"fn":1677,"description":1678,"org":1679,"tags":1680,"stars":20,"repoUrl":21,"updatedAt":1688},"cudaq-guide","develop quantum applications with CUDA-Q","CUDA-Q onboarding guide for installation, test programs, GPU simulation, QPU hardware, and quantum applications.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1681,1682,1685],{"name":9,"slug":8,"type":13},{"name":1683,"slug":1684,"type":13},"Quantum Computing","quantum-computing",{"name":1686,"slug":1687,"type":13},"Simulation","simulation","2026-07-14T05:26:58.898253",305]