[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-earth2studio-deterministic-forecast":3,"mdc--c7oi5z-key":34,"related-org-nvidia-earth2studio-deterministic-forecast":1008,"related-repo-nvidia-earth2studio-deterministic-forecast":1167},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":29,"sourceUrl":32,"mdContent":33},"earth2studio-deterministic-forecast","build Earth2Studio deterministic forecast scripts","Build deterministic forecast scripts with Earth2Studio (model, data source, IO, inference). Do NOT use for ensemble, diagnostics, data-only fetch, or install.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,17,20],{"name":13,"slug":14,"type":15},"Forecasting","forecasting","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Simulation","simulation",{"name":21,"slug":22,"type":15},"Python","python",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-14T05:28:32.052907","Apache-2.0",281,[],{"repoUrl":24,"stars":23,"forks":27,"topics":30,"description":31},[],"AI agent skills published by NVIDIA","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fearth2studio-deterministic-forecast","---\nname: earth2studio-deterministic-forecast\nversion: 0.16.0\nlicense: Apache-2.0\nmetadata:\n  author: NVIDIA Earth-2 Team\n  tags:\n    - earth2studio\n    - earth2\n    - python\n    - inference\n    - forecast\n    - deterministic\ndescription: >\n  Build deterministic forecast scripts with Earth2Studio (model, data source,\n  IO, inference). Do NOT use for ensemble, diagnostics, data-only fetch, or\n  install.\n---\n\n# Earth2Studio Deterministic Forecast Skill\n\nGuide users through building deterministic (single-member) weather forecast\ninference scripts using `earth2studio.run.deterministic`.\n\n## Prerequisites\n\n- Earth2Studio installed with CUDA-capable GPU\n- Python 3.10+, network access for model weights and data\n\n## Live Doc References\n\nFetch relevant docs to verify current APIs before recommending components:\n\n| Component | URL |\n|-----------|-----|\n| Prognostic models | \u003Chttps:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fmodels_px.html> |\n| Data sources (analysis) | \u003Chttps:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fdatasources_analysis.html> |\n| Data sources (forecast) | \u003Chttps:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fdatasources_forecast.html> |\n| IO backends | \u003Chttps:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fio.html> |\n| `run.deterministic` | \u003Chttps:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fearth2studio\u002Fblob\u002Fmain\u002Fearth2studio\u002Frun.py> |\n\n## Workflow\n\n### 1. Gather Requirements (skip what's already provided)\n\n- Time horizon (hours\u002Fdays\u002Fweeks)\n- Variables of interest (t2m, wind, geopotential, etc.)\n- Region (global or specific like CONUS)\n- GPU\u002FVRAM available\n\n### 2. Select Model\n\nFetch prognostic models page. Filter by time horizon, region, VRAM. Note model's:\n- Input variables (`input_coords[\"variable\"]`)\n- Time step size (`output_coords[\"lead_time\"]`)\n\n### 3. Select Data Source\n\nData source must provide all model input variables. Verify via lexicon at\n`earth2studio\u002Flexicon\u002F\u003Csource>.py`. Common pairings: Global models → GFS\u002FARCO\u002FIFS;\nRegional → HRRR.\n\n### 4. Select IO Backend\n\nDefault: `ZarrBackend`. Use `NetCDF4Backend` for legacy tools, `XarrayBackend`\nfor in-memory\u002Fsmall runs.\n\n### 5. Calculate nsteps\n\n`nsteps = forecast_hours \u002F model_step_hours`\n\nExample: 5-day forecast with 6h step → `nsteps = 120 \u002F 6 = 20`\n\n### 6. Decide: output_coords Filtering\n\n- **Filter variables** (`output_coords`) when user requests specific variables (e.g., \"t2m and wind\") - reduces output size\n- **Save all variables** (omit `output_coords`) when user says \"all variables\" or doesn't specify - preserves full model output\n\n### 7. Generate Script\n\n```python\nfrom collections import OrderedDict\nimport numpy as np\nimport torch\nfrom earth2studio.models.px import \u003CModelClass>\nfrom earth2studio.data import \u003CDataSourceClass>\nfrom earth2studio.io import \u003CIOBackendClass>\nfrom earth2studio.run import deterministic\n\nmodel = \u003CModelClass>.load_model(\u003CModelClass>.load_default_package())\ndata = \u003CDataSourceClass>()\nio = \u003CIOBackendClass>(\"\u003Coutput_path>\")\n\n# Include output_coords ONLY if user requested specific variables\noutput_coords = OrderedDict({\"variable\": np.array([\"t2m\", \"u10m\"])})\n\nio = deterministic(\n    time=[\"YYYY-MM-DDTHH:MM:SS\"],\n    nsteps=\u003CN>,\n    prognostic=model,\n    data=data,\n    io=io,\n    output_coords=output_coords,  # omit if saving all variables\n    device=torch.device(\"cuda\"),\n)\n```\n\n### 8. Manual Loop Alternative\n\nWhen user explicitly requests manual implementation (NOT using `earth2studio.run.deterministic`), follow this checklist in order:\n\n1. **fetch_data** - Get initial conditions: `x, coords = fetch_data(data, time, model.input_coords, device)`\n2. **Setup total_coords** - Build coordinate arrays for time and lead_time dimensions\n3. **io.add_array** - Initialize IO backend with total_coords before loop\n4. **create_iterator** - Create prognostic iterator: `model_iter = model.create_iterator(x, coords)`\n5. **Loop through nsteps** - `for step, (x, coords) in enumerate(model_iter): if step >= nsteps: break`\n6. **map_coords** - Filter output variables if needed: `x_out, coords_out = map_coords(x, coords, output_coords)`\n7. **split_coords** - Prepare for IO write: `x_out, coords_out = split_coords(x_out, coords_out)`\n8. **io.write** - Write each step to backend\n\n### 9. Explain Next Steps\n\n- How to change forecast time or run multiple initializations\n- How to read output (`xr.open_zarr(...)`)\n- Point to diagnostic workflow for post-processing\n\n## Ownership\n\n**Owns:** Model selection, data source compatibility, IO backend selection,\nnsteps calculation, generating `earth2studio.run.deterministic` scripts.\n\n**Does not own:** Ensemble workflows, diagnostics, data-only fetch, installation,\nmodel training.\n\n## Troubleshooting\n\nSee `references\u002Ftroubleshooting.md` for common errors and solutions.\n\n## Reminders\n\n- **Always fetch live docs** before recommending models or data sources - APIs change between releases\n- **Verify lexicon compatibility** - Model input variables must exist in data source's VOCAB\n- **Use `load_default_package()`** - This is the standard pattern for loading model weights\n- **Time format is ISO 8601** - Use `\"YYYY-MM-DDTHH:MM:SS\"` format for the `time` argument\n- **Wind speed needs both components** - If user asks for \"wind speed\", include both `u10m` and `v10m`\n- **nsteps is integer division** - `nsteps = total_hours \u002F\u002F model_step_hours`\n- **ZarrBackend is the default** - Only suggest alternatives if user has specific requirements\n- **GPU is required** - All prognostic models require CUDA; CPU inference is not supported\n",{"data":35,"body":45},{"name":4,"version":36,"license":26,"metadata":37,"description":6},"0.16.0",{"author":38,"tags":39},"NVIDIA Earth-2 Team",[40,41,22,42,43,44],"earth2studio","earth2","inference","forecast","deterministic",{"type":46,"children":47},"root",[48,57,72,79,94,100,105,224,230,237,260,266,271,299,305,318,324,353,359,368,379,385,424,430,657,663,675,789,795,820,826,843,853,859,872,878,1002],{"type":49,"tag":50,"props":51,"children":53},"element","h1",{"id":52},"earth2studio-deterministic-forecast-skill",[54],{"type":55,"value":56},"text","Earth2Studio Deterministic Forecast Skill",{"type":49,"tag":58,"props":59,"children":60},"p",{},[61,63,70],{"type":55,"value":62},"Guide users through building deterministic (single-member) weather forecast\ninference scripts using ",{"type":49,"tag":64,"props":65,"children":67},"code",{"className":66},[],[68],{"type":55,"value":69},"earth2studio.run.deterministic",{"type":55,"value":71},".",{"type":49,"tag":73,"props":74,"children":76},"h2",{"id":75},"prerequisites",[77],{"type":55,"value":78},"Prerequisites",{"type":49,"tag":80,"props":81,"children":82},"ul",{},[83,89],{"type":49,"tag":84,"props":85,"children":86},"li",{},[87],{"type":55,"value":88},"Earth2Studio installed with CUDA-capable GPU",{"type":49,"tag":84,"props":90,"children":91},{},[92],{"type":55,"value":93},"Python 3.10+, network access for model weights and data",{"type":49,"tag":73,"props":95,"children":97},{"id":96},"live-doc-references",[98],{"type":55,"value":99},"Live Doc References",{"type":49,"tag":58,"props":101,"children":102},{},[103],{"type":55,"value":104},"Fetch relevant docs to verify current APIs before recommending components:",{"type":49,"tag":106,"props":107,"children":108},"table",{},[109,128],{"type":49,"tag":110,"props":111,"children":112},"thead",{},[113],{"type":49,"tag":114,"props":115,"children":116},"tr",{},[117,123],{"type":49,"tag":118,"props":119,"children":120},"th",{},[121],{"type":55,"value":122},"Component",{"type":49,"tag":118,"props":124,"children":125},{},[126],{"type":55,"value":127},"URL",{"type":49,"tag":129,"props":130,"children":131},"tbody",{},[132,152,169,186,203],{"type":49,"tag":114,"props":133,"children":134},{},[135,141],{"type":49,"tag":136,"props":137,"children":138},"td",{},[139],{"type":55,"value":140},"Prognostic models",{"type":49,"tag":136,"props":142,"children":143},{},[144],{"type":49,"tag":145,"props":146,"children":150},"a",{"href":147,"rel":148},"https:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fmodels_px.html",[149],"nofollow",[151],{"type":55,"value":147},{"type":49,"tag":114,"props":153,"children":154},{},[155,160],{"type":49,"tag":136,"props":156,"children":157},{},[158],{"type":55,"value":159},"Data sources (analysis)",{"type":49,"tag":136,"props":161,"children":162},{},[163],{"type":49,"tag":145,"props":164,"children":167},{"href":165,"rel":166},"https:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fdatasources_analysis.html",[149],[168],{"type":55,"value":165},{"type":49,"tag":114,"props":170,"children":171},{},[172,177],{"type":49,"tag":136,"props":173,"children":174},{},[175],{"type":55,"value":176},"Data sources (forecast)",{"type":49,"tag":136,"props":178,"children":179},{},[180],{"type":49,"tag":145,"props":181,"children":184},{"href":182,"rel":183},"https:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fdatasources_forecast.html",[149],[185],{"type":55,"value":182},{"type":49,"tag":114,"props":187,"children":188},{},[189,194],{"type":49,"tag":136,"props":190,"children":191},{},[192],{"type":55,"value":193},"IO backends",{"type":49,"tag":136,"props":195,"children":196},{},[197],{"type":49,"tag":145,"props":198,"children":201},{"href":199,"rel":200},"https:\u002F\u002Fnvidia.github.io\u002Fearth2studio\u002Fmodules\u002Fio.html",[149],[202],{"type":55,"value":199},{"type":49,"tag":114,"props":204,"children":205},{},[206,215],{"type":49,"tag":136,"props":207,"children":208},{},[209],{"type":49,"tag":64,"props":210,"children":212},{"className":211},[],[213],{"type":55,"value":214},"run.deterministic",{"type":49,"tag":136,"props":216,"children":217},{},[218],{"type":49,"tag":145,"props":219,"children":222},{"href":220,"rel":221},"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fearth2studio\u002Fblob\u002Fmain\u002Fearth2studio\u002Frun.py",[149],[223],{"type":55,"value":220},{"type":49,"tag":73,"props":225,"children":227},{"id":226},"workflow",[228],{"type":55,"value":229},"Workflow",{"type":49,"tag":231,"props":232,"children":234},"h3",{"id":233},"_1-gather-requirements-skip-whats-already-provided",[235],{"type":55,"value":236},"1. Gather Requirements (skip what's already provided)",{"type":49,"tag":80,"props":238,"children":239},{},[240,245,250,255],{"type":49,"tag":84,"props":241,"children":242},{},[243],{"type":55,"value":244},"Time horizon (hours\u002Fdays\u002Fweeks)",{"type":49,"tag":84,"props":246,"children":247},{},[248],{"type":55,"value":249},"Variables of interest (t2m, wind, geopotential, etc.)",{"type":49,"tag":84,"props":251,"children":252},{},[253],{"type":55,"value":254},"Region (global or specific like CONUS)",{"type":49,"tag":84,"props":256,"children":257},{},[258],{"type":55,"value":259},"GPU\u002FVRAM available",{"type":49,"tag":231,"props":261,"children":263},{"id":262},"_2-select-model",[264],{"type":55,"value":265},"2. Select Model",{"type":49,"tag":58,"props":267,"children":268},{},[269],{"type":55,"value":270},"Fetch prognostic models page. Filter by time horizon, region, VRAM. Note model's:",{"type":49,"tag":80,"props":272,"children":273},{},[274,287],{"type":49,"tag":84,"props":275,"children":276},{},[277,279,285],{"type":55,"value":278},"Input variables (",{"type":49,"tag":64,"props":280,"children":282},{"className":281},[],[283],{"type":55,"value":284},"input_coords[\"variable\"]",{"type":55,"value":286},")",{"type":49,"tag":84,"props":288,"children":289},{},[290,292,298],{"type":55,"value":291},"Time step size (",{"type":49,"tag":64,"props":293,"children":295},{"className":294},[],[296],{"type":55,"value":297},"output_coords[\"lead_time\"]",{"type":55,"value":286},{"type":49,"tag":231,"props":300,"children":302},{"id":301},"_3-select-data-source",[303],{"type":55,"value":304},"3. Select Data Source",{"type":49,"tag":58,"props":306,"children":307},{},[308,310,316],{"type":55,"value":309},"Data source must provide all model input variables. Verify via lexicon at\n",{"type":49,"tag":64,"props":311,"children":313},{"className":312},[],[314],{"type":55,"value":315},"earth2studio\u002Flexicon\u002F\u003Csource>.py",{"type":55,"value":317},". Common pairings: Global models → GFS\u002FARCO\u002FIFS;\nRegional → HRRR.",{"type":49,"tag":231,"props":319,"children":321},{"id":320},"_4-select-io-backend",[322],{"type":55,"value":323},"4. Select IO Backend",{"type":49,"tag":58,"props":325,"children":326},{},[327,329,335,337,343,345,351],{"type":55,"value":328},"Default: ",{"type":49,"tag":64,"props":330,"children":332},{"className":331},[],[333],{"type":55,"value":334},"ZarrBackend",{"type":55,"value":336},". Use ",{"type":49,"tag":64,"props":338,"children":340},{"className":339},[],[341],{"type":55,"value":342},"NetCDF4Backend",{"type":55,"value":344}," for legacy tools, ",{"type":49,"tag":64,"props":346,"children":348},{"className":347},[],[349],{"type":55,"value":350},"XarrayBackend",{"type":55,"value":352},"\nfor in-memory\u002Fsmall runs.",{"type":49,"tag":231,"props":354,"children":356},{"id":355},"_5-calculate-nsteps",[357],{"type":55,"value":358},"5. Calculate nsteps",{"type":49,"tag":58,"props":360,"children":361},{},[362],{"type":49,"tag":64,"props":363,"children":365},{"className":364},[],[366],{"type":55,"value":367},"nsteps = forecast_hours \u002F model_step_hours",{"type":49,"tag":58,"props":369,"children":370},{},[371,373],{"type":55,"value":372},"Example: 5-day forecast with 6h step → ",{"type":49,"tag":64,"props":374,"children":376},{"className":375},[],[377],{"type":55,"value":378},"nsteps = 120 \u002F 6 = 20",{"type":49,"tag":231,"props":380,"children":382},{"id":381},"_6-decide-output_coords-filtering",[383],{"type":55,"value":384},"6. Decide: output_coords Filtering",{"type":49,"tag":80,"props":386,"children":387},{},[388,407],{"type":49,"tag":84,"props":389,"children":390},{},[391,397,399,405],{"type":49,"tag":392,"props":393,"children":394},"strong",{},[395],{"type":55,"value":396},"Filter variables",{"type":55,"value":398}," (",{"type":49,"tag":64,"props":400,"children":402},{"className":401},[],[403],{"type":55,"value":404},"output_coords",{"type":55,"value":406},") when user requests specific variables (e.g., \"t2m and wind\") - reduces output size",{"type":49,"tag":84,"props":408,"children":409},{},[410,415,417,422],{"type":49,"tag":392,"props":411,"children":412},{},[413],{"type":55,"value":414},"Save all variables",{"type":55,"value":416}," (omit ",{"type":49,"tag":64,"props":418,"children":420},{"className":419},[],[421],{"type":55,"value":404},{"type":55,"value":423},") when user says \"all variables\" or doesn't specify - preserves full model output",{"type":49,"tag":231,"props":425,"children":427},{"id":426},"_7-generate-script",[428],{"type":55,"value":429},"7. Generate Script",{"type":49,"tag":431,"props":432,"children":436},"pre",{"className":433,"code":434,"language":22,"meta":435,"style":435},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","from collections import OrderedDict\nimport numpy as np\nimport torch\nfrom earth2studio.models.px import \u003CModelClass>\nfrom earth2studio.data import \u003CDataSourceClass>\nfrom earth2studio.io import \u003CIOBackendClass>\nfrom earth2studio.run import deterministic\n\nmodel = \u003CModelClass>.load_model(\u003CModelClass>.load_default_package())\ndata = \u003CDataSourceClass>()\nio = \u003CIOBackendClass>(\"\u003Coutput_path>\")\n\n# Include output_coords ONLY if user requested specific variables\noutput_coords = OrderedDict({\"variable\": np.array([\"t2m\", \"u10m\"])})\n\nio = deterministic(\n    time=[\"YYYY-MM-DDTHH:MM:SS\"],\n    nsteps=\u003CN>,\n    prognostic=model,\n    data=data,\n    io=io,\n    output_coords=output_coords,  # omit if saving all variables\n    device=torch.device(\"cuda\"),\n)\n","",[437],{"type":49,"tag":64,"props":438,"children":439},{"__ignoreMap":435},[440,451,460,469,478,487,496,505,515,524,533,542,550,559,568,576,585,594,603,612,621,630,639,648],{"type":49,"tag":441,"props":442,"children":445},"span",{"class":443,"line":444},"line",1,[446],{"type":49,"tag":441,"props":447,"children":448},{},[449],{"type":55,"value":450},"from collections import OrderedDict\n",{"type":49,"tag":441,"props":452,"children":454},{"class":443,"line":453},2,[455],{"type":49,"tag":441,"props":456,"children":457},{},[458],{"type":55,"value":459},"import numpy as np\n",{"type":49,"tag":441,"props":461,"children":463},{"class":443,"line":462},3,[464],{"type":49,"tag":441,"props":465,"children":466},{},[467],{"type":55,"value":468},"import torch\n",{"type":49,"tag":441,"props":470,"children":472},{"class":443,"line":471},4,[473],{"type":49,"tag":441,"props":474,"children":475},{},[476],{"type":55,"value":477},"from earth2studio.models.px import \u003CModelClass>\n",{"type":49,"tag":441,"props":479,"children":481},{"class":443,"line":480},5,[482],{"type":49,"tag":441,"props":483,"children":484},{},[485],{"type":55,"value":486},"from earth2studio.data import \u003CDataSourceClass>\n",{"type":49,"tag":441,"props":488,"children":490},{"class":443,"line":489},6,[491],{"type":49,"tag":441,"props":492,"children":493},{},[494],{"type":55,"value":495},"from earth2studio.io import \u003CIOBackendClass>\n",{"type":49,"tag":441,"props":497,"children":499},{"class":443,"line":498},7,[500],{"type":49,"tag":441,"props":501,"children":502},{},[503],{"type":55,"value":504},"from earth2studio.run import deterministic\n",{"type":49,"tag":441,"props":506,"children":508},{"class":443,"line":507},8,[509],{"type":49,"tag":441,"props":510,"children":512},{"emptyLinePlaceholder":511},true,[513],{"type":55,"value":514},"\n",{"type":49,"tag":441,"props":516,"children":518},{"class":443,"line":517},9,[519],{"type":49,"tag":441,"props":520,"children":521},{},[522],{"type":55,"value":523},"model = \u003CModelClass>.load_model(\u003CModelClass>.load_default_package())\n",{"type":49,"tag":441,"props":525,"children":527},{"class":443,"line":526},10,[528],{"type":49,"tag":441,"props":529,"children":530},{},[531],{"type":55,"value":532},"data = \u003CDataSourceClass>()\n",{"type":49,"tag":441,"props":534,"children":536},{"class":443,"line":535},11,[537],{"type":49,"tag":441,"props":538,"children":539},{},[540],{"type":55,"value":541},"io = \u003CIOBackendClass>(\"\u003Coutput_path>\")\n",{"type":49,"tag":441,"props":543,"children":545},{"class":443,"line":544},12,[546],{"type":49,"tag":441,"props":547,"children":548},{"emptyLinePlaceholder":511},[549],{"type":55,"value":514},{"type":49,"tag":441,"props":551,"children":553},{"class":443,"line":552},13,[554],{"type":49,"tag":441,"props":555,"children":556},{},[557],{"type":55,"value":558},"# Include output_coords ONLY if user requested specific variables\n",{"type":49,"tag":441,"props":560,"children":562},{"class":443,"line":561},14,[563],{"type":49,"tag":441,"props":564,"children":565},{},[566],{"type":55,"value":567},"output_coords = OrderedDict({\"variable\": np.array([\"t2m\", \"u10m\"])})\n",{"type":49,"tag":441,"props":569,"children":571},{"class":443,"line":570},15,[572],{"type":49,"tag":441,"props":573,"children":574},{"emptyLinePlaceholder":511},[575],{"type":55,"value":514},{"type":49,"tag":441,"props":577,"children":579},{"class":443,"line":578},16,[580],{"type":49,"tag":441,"props":581,"children":582},{},[583],{"type":55,"value":584},"io = deterministic(\n",{"type":49,"tag":441,"props":586,"children":588},{"class":443,"line":587},17,[589],{"type":49,"tag":441,"props":590,"children":591},{},[592],{"type":55,"value":593},"    time=[\"YYYY-MM-DDTHH:MM:SS\"],\n",{"type":49,"tag":441,"props":595,"children":597},{"class":443,"line":596},18,[598],{"type":49,"tag":441,"props":599,"children":600},{},[601],{"type":55,"value":602},"    nsteps=\u003CN>,\n",{"type":49,"tag":441,"props":604,"children":606},{"class":443,"line":605},19,[607],{"type":49,"tag":441,"props":608,"children":609},{},[610],{"type":55,"value":611},"    prognostic=model,\n",{"type":49,"tag":441,"props":613,"children":615},{"class":443,"line":614},20,[616],{"type":49,"tag":441,"props":617,"children":618},{},[619],{"type":55,"value":620},"    data=data,\n",{"type":49,"tag":441,"props":622,"children":624},{"class":443,"line":623},21,[625],{"type":49,"tag":441,"props":626,"children":627},{},[628],{"type":55,"value":629},"    io=io,\n",{"type":49,"tag":441,"props":631,"children":633},{"class":443,"line":632},22,[634],{"type":49,"tag":441,"props":635,"children":636},{},[637],{"type":55,"value":638},"    output_coords=output_coords,  # omit if saving all variables\n",{"type":49,"tag":441,"props":640,"children":642},{"class":443,"line":641},23,[643],{"type":49,"tag":441,"props":644,"children":645},{},[646],{"type":55,"value":647},"    device=torch.device(\"cuda\"),\n",{"type":49,"tag":441,"props":649,"children":651},{"class":443,"line":650},24,[652],{"type":49,"tag":441,"props":653,"children":654},{},[655],{"type":55,"value":656},")\n",{"type":49,"tag":231,"props":658,"children":660},{"id":659},"_8-manual-loop-alternative",[661],{"type":55,"value":662},"8. Manual Loop Alternative",{"type":49,"tag":58,"props":664,"children":665},{},[666,668,673],{"type":55,"value":667},"When user explicitly requests manual implementation (NOT using ",{"type":49,"tag":64,"props":669,"children":671},{"className":670},[],[672],{"type":55,"value":69},{"type":55,"value":674},"), follow this checklist in order:",{"type":49,"tag":676,"props":677,"children":678},"ol",{},[679,695,705,715,731,747,763,779],{"type":49,"tag":84,"props":680,"children":681},{},[682,687,689],{"type":49,"tag":392,"props":683,"children":684},{},[685],{"type":55,"value":686},"fetch_data",{"type":55,"value":688}," - Get initial conditions: ",{"type":49,"tag":64,"props":690,"children":692},{"className":691},[],[693],{"type":55,"value":694},"x, coords = fetch_data(data, time, model.input_coords, device)",{"type":49,"tag":84,"props":696,"children":697},{},[698,703],{"type":49,"tag":392,"props":699,"children":700},{},[701],{"type":55,"value":702},"Setup total_coords",{"type":55,"value":704}," - Build coordinate arrays for time and lead_time dimensions",{"type":49,"tag":84,"props":706,"children":707},{},[708,713],{"type":49,"tag":392,"props":709,"children":710},{},[711],{"type":55,"value":712},"io.add_array",{"type":55,"value":714}," - Initialize IO backend with total_coords before loop",{"type":49,"tag":84,"props":716,"children":717},{},[718,723,725],{"type":49,"tag":392,"props":719,"children":720},{},[721],{"type":55,"value":722},"create_iterator",{"type":55,"value":724}," - Create prognostic iterator: ",{"type":49,"tag":64,"props":726,"children":728},{"className":727},[],[729],{"type":55,"value":730},"model_iter = model.create_iterator(x, coords)",{"type":49,"tag":84,"props":732,"children":733},{},[734,739,741],{"type":49,"tag":392,"props":735,"children":736},{},[737],{"type":55,"value":738},"Loop through nsteps",{"type":55,"value":740}," - ",{"type":49,"tag":64,"props":742,"children":744},{"className":743},[],[745],{"type":55,"value":746},"for step, (x, coords) in enumerate(model_iter): if step >= nsteps: break",{"type":49,"tag":84,"props":748,"children":749},{},[750,755,757],{"type":49,"tag":392,"props":751,"children":752},{},[753],{"type":55,"value":754},"map_coords",{"type":55,"value":756}," - Filter output variables if needed: ",{"type":49,"tag":64,"props":758,"children":760},{"className":759},[],[761],{"type":55,"value":762},"x_out, coords_out = map_coords(x, coords, output_coords)",{"type":49,"tag":84,"props":764,"children":765},{},[766,771,773],{"type":49,"tag":392,"props":767,"children":768},{},[769],{"type":55,"value":770},"split_coords",{"type":55,"value":772}," - Prepare for IO write: ",{"type":49,"tag":64,"props":774,"children":776},{"className":775},[],[777],{"type":55,"value":778},"x_out, coords_out = split_coords(x_out, coords_out)",{"type":49,"tag":84,"props":780,"children":781},{},[782,787],{"type":49,"tag":392,"props":783,"children":784},{},[785],{"type":55,"value":786},"io.write",{"type":55,"value":788}," - Write each step to backend",{"type":49,"tag":231,"props":790,"children":792},{"id":791},"_9-explain-next-steps",[793],{"type":55,"value":794},"9. Explain Next Steps",{"type":49,"tag":80,"props":796,"children":797},{},[798,803,815],{"type":49,"tag":84,"props":799,"children":800},{},[801],{"type":55,"value":802},"How to change forecast time or run multiple initializations",{"type":49,"tag":84,"props":804,"children":805},{},[806,808,814],{"type":55,"value":807},"How to read output (",{"type":49,"tag":64,"props":809,"children":811},{"className":810},[],[812],{"type":55,"value":813},"xr.open_zarr(...)",{"type":55,"value":286},{"type":49,"tag":84,"props":816,"children":817},{},[818],{"type":55,"value":819},"Point to diagnostic workflow for post-processing",{"type":49,"tag":73,"props":821,"children":823},{"id":822},"ownership",[824],{"type":55,"value":825},"Ownership",{"type":49,"tag":58,"props":827,"children":828},{},[829,834,836,841],{"type":49,"tag":392,"props":830,"children":831},{},[832],{"type":55,"value":833},"Owns:",{"type":55,"value":835}," Model selection, data source compatibility, IO backend selection,\nnsteps calculation, generating ",{"type":49,"tag":64,"props":837,"children":839},{"className":838},[],[840],{"type":55,"value":69},{"type":55,"value":842}," scripts.",{"type":49,"tag":58,"props":844,"children":845},{},[846,851],{"type":49,"tag":392,"props":847,"children":848},{},[849],{"type":55,"value":850},"Does not own:",{"type":55,"value":852}," Ensemble workflows, diagnostics, data-only fetch, installation,\nmodel training.",{"type":49,"tag":73,"props":854,"children":856},{"id":855},"troubleshooting",[857],{"type":55,"value":858},"Troubleshooting",{"type":49,"tag":58,"props":860,"children":861},{},[862,864,870],{"type":55,"value":863},"See ",{"type":49,"tag":64,"props":865,"children":867},{"className":866},[],[868],{"type":55,"value":869},"references\u002Ftroubleshooting.md",{"type":55,"value":871}," for common errors and solutions.",{"type":49,"tag":73,"props":873,"children":875},{"id":874},"reminders",[876],{"type":55,"value":877},"Reminders",{"type":49,"tag":80,"props":879,"children":880},{},[881,891,901,917,943,967,982,992],{"type":49,"tag":84,"props":882,"children":883},{},[884,889],{"type":49,"tag":392,"props":885,"children":886},{},[887],{"type":55,"value":888},"Always fetch live docs",{"type":55,"value":890}," before recommending models or data sources - APIs change between releases",{"type":49,"tag":84,"props":892,"children":893},{},[894,899],{"type":49,"tag":392,"props":895,"children":896},{},[897],{"type":55,"value":898},"Verify lexicon compatibility",{"type":55,"value":900}," - Model input variables must exist in data source's VOCAB",{"type":49,"tag":84,"props":902,"children":903},{},[904,915],{"type":49,"tag":392,"props":905,"children":906},{},[907,909],{"type":55,"value":908},"Use ",{"type":49,"tag":64,"props":910,"children":912},{"className":911},[],[913],{"type":55,"value":914},"load_default_package()",{"type":55,"value":916}," - This is the standard pattern for loading model weights",{"type":49,"tag":84,"props":918,"children":919},{},[920,925,927,933,935,941],{"type":49,"tag":392,"props":921,"children":922},{},[923],{"type":55,"value":924},"Time format is ISO 8601",{"type":55,"value":926}," - Use ",{"type":49,"tag":64,"props":928,"children":930},{"className":929},[],[931],{"type":55,"value":932},"\"YYYY-MM-DDTHH:MM:SS\"",{"type":55,"value":934}," format for the ",{"type":49,"tag":64,"props":936,"children":938},{"className":937},[],[939],{"type":55,"value":940},"time",{"type":55,"value":942}," argument",{"type":49,"tag":84,"props":944,"children":945},{},[946,951,953,959,961],{"type":49,"tag":392,"props":947,"children":948},{},[949],{"type":55,"value":950},"Wind speed needs both components",{"type":55,"value":952}," - If user asks for \"wind speed\", include both ",{"type":49,"tag":64,"props":954,"children":956},{"className":955},[],[957],{"type":55,"value":958},"u10m",{"type":55,"value":960}," and ",{"type":49,"tag":64,"props":962,"children":964},{"className":963},[],[965],{"type":55,"value":966},"v10m",{"type":49,"tag":84,"props":968,"children":969},{},[970,975,976],{"type":49,"tag":392,"props":971,"children":972},{},[973],{"type":55,"value":974},"nsteps is integer division",{"type":55,"value":740},{"type":49,"tag":64,"props":977,"children":979},{"className":978},[],[980],{"type":55,"value":981},"nsteps = total_hours \u002F\u002F model_step_hours",{"type":49,"tag":84,"props":983,"children":984},{},[985,990],{"type":49,"tag":392,"props":986,"children":987},{},[988],{"type":55,"value":989},"ZarrBackend is the default",{"type":55,"value":991}," - Only suggest alternatives if user has specific requirements",{"type":49,"tag":84,"props":993,"children":994},{},[995,1000],{"type":49,"tag":392,"props":996,"children":997},{},[998],{"type":55,"value":999},"GPU is required",{"type":55,"value":1001}," - All prognostic models require CUDA; CPU inference is not supported",{"type":49,"tag":1003,"props":1004,"children":1005},"style",{},[1006],{"type":55,"value":1007},"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":1009,"total":1166},[1010,1028,1044,1055,1067,1081,1094,1108,1121,1132,1146,1155],{"slug":1011,"name":1011,"fn":1012,"description":1013,"org":1014,"tags":1015,"stars":1025,"repoUrl":1026,"updatedAt":1027},"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},[1016,1019,1022],{"name":1017,"slug":1018,"type":15},"Documentation","documentation",{"name":1020,"slug":1021,"type":15},"MCP","mcp",{"name":1023,"slug":1024,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1029,"name":1029,"fn":1030,"description":1031,"org":1032,"tags":1033,"stars":1041,"repoUrl":1042,"updatedAt":1043},"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},[1034,1037,1040],{"name":1035,"slug":1036,"type":15},"Containers","containers",{"name":1038,"slug":1039,"type":15},"Deployment","deployment",{"name":21,"slug":22,"type":15},17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1045,"name":1045,"fn":1046,"description":1047,"org":1048,"tags":1049,"stars":1041,"repoUrl":1042,"updatedAt":1054},"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},[1050,1053],{"name":1051,"slug":1052,"type":15},"CI\u002FCD","ci-cd",{"name":1038,"slug":1039,"type":15},"2026-07-14T05:25:59.97109",{"slug":1056,"name":1056,"fn":1057,"description":1058,"org":1059,"tags":1060,"stars":1041,"repoUrl":1042,"updatedAt":1066},"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},[1061,1062,1063],{"name":1051,"slug":1052,"type":15},{"name":1038,"slug":1039,"type":15},{"name":1064,"slug":1065,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1068,"name":1068,"fn":1069,"description":1070,"org":1071,"tags":1072,"stars":1041,"repoUrl":1042,"updatedAt":1080},"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},[1073,1076,1077],{"name":1074,"slug":1075,"type":15},"Debugging","debugging",{"name":1064,"slug":1065,"type":15},{"name":1078,"slug":1079,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1082,"name":1082,"fn":1083,"description":1084,"org":1085,"tags":1086,"stars":1041,"repoUrl":1042,"updatedAt":1093},"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},[1087,1090],{"name":1088,"slug":1089,"type":15},"Best Practices","best-practices",{"name":1091,"slug":1092,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1095,"name":1095,"fn":1096,"description":1097,"org":1098,"tags":1099,"stars":1041,"repoUrl":1042,"updatedAt":1107},"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},[1100,1103,1106],{"name":1101,"slug":1102,"type":15},"Machine Learning","machine-learning",{"name":1104,"slug":1105,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1109,"name":1109,"fn":1110,"description":1111,"org":1112,"tags":1113,"stars":1041,"repoUrl":1042,"updatedAt":1120},"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},[1114,1117],{"name":1115,"slug":1116,"type":15},"QA","qa",{"name":1118,"slug":1119,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":1122,"name":1122,"fn":1123,"description":1124,"org":1125,"tags":1126,"stars":1041,"repoUrl":1042,"updatedAt":1131},"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},[1127,1128],{"name":1038,"slug":1039,"type":15},{"name":1129,"slug":1130,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":1133,"name":1133,"fn":1134,"description":1135,"org":1136,"tags":1137,"stars":1041,"repoUrl":1042,"updatedAt":1145},"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},[1138,1141,1142],{"name":1139,"slug":1140,"type":15},"Code Review","code-review",{"name":1064,"slug":1065,"type":15},{"name":1143,"slug":1144,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1147,"name":1147,"fn":1148,"description":1149,"org":1150,"tags":1151,"stars":1041,"repoUrl":1042,"updatedAt":1154},"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},[1152,1153],{"name":1115,"slug":1116,"type":15},{"name":1118,"slug":1119,"type":15},"2026-07-14T05:25:54.928983",{"slug":1156,"name":1156,"fn":1157,"description":1158,"org":1159,"tags":1160,"stars":1041,"repoUrl":1042,"updatedAt":1165},"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},[1161,1164],{"name":1162,"slug":1163,"type":15},"Automation","automation",{"name":1051,"slug":1052,"type":15},"2026-07-30T05:29:03.275638",496,{"items":1168,"total":1262},[1169,1186,1196,1210,1220,1235,1250],{"slug":1170,"name":1170,"fn":1171,"description":1172,"org":1173,"tags":1174,"stars":23,"repoUrl":24,"updatedAt":1185},"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},[1175,1178,1181,1182],{"name":1176,"slug":1177,"type":15},"Data Analysis","data-analysis",{"name":1179,"slug":1180,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":1183,"slug":1184,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":1187,"name":1187,"fn":1188,"description":1189,"org":1190,"tags":1191,"stars":23,"repoUrl":24,"updatedAt":1195},"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},[1192,1193,1194],{"name":1038,"slug":1039,"type":15},{"name":1129,"slug":1130,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":1197,"name":1197,"fn":1198,"description":1199,"org":1200,"tags":1201,"stars":23,"repoUrl":24,"updatedAt":1209},"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},[1202,1205,1206],{"name":1203,"slug":1204,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":1207,"slug":1208,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":1211,"name":1211,"fn":1212,"description":1213,"org":1214,"tags":1215,"stars":23,"repoUrl":24,"updatedAt":1219},"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},[1216,1217,1218],{"name":1176,"slug":1177,"type":15},{"name":9,"slug":8,"type":15},{"name":1118,"slug":1119,"type":15},"2026-07-17T05:29:03.913266",{"slug":1221,"name":1221,"fn":1222,"description":1223,"org":1224,"tags":1225,"stars":23,"repoUrl":24,"updatedAt":1234},"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},[1226,1227,1230,1231],{"name":1162,"slug":1163,"type":15},{"name":1228,"slug":1229,"type":15},"Imaging","imaging",{"name":9,"slug":8,"type":15},{"name":1232,"slug":1233,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":1236,"name":1236,"fn":1237,"description":1238,"org":1239,"tags":1240,"stars":23,"repoUrl":24,"updatedAt":1249},"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},[1241,1242,1245,1246],{"name":1038,"slug":1039,"type":15},{"name":1243,"slug":1244,"type":15},"Docker","docker",{"name":9,"slug":8,"type":15},{"name":1247,"slug":1248,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":1251,"name":1251,"fn":1252,"description":1253,"org":1254,"tags":1255,"stars":23,"repoUrl":24,"updatedAt":1261},"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},[1256,1257,1260],{"name":9,"slug":8,"type":15},{"name":1258,"slug":1259,"type":15},"Quantum Computing","quantum-computing",{"name":18,"slug":19,"type":15},"2026-07-14T05:26:58.898253",305]