[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-cuopt-routing-api-python":3,"mdc--t7mxz2-key":34,"related-repo-nvidia-cuopt-routing-api-python":799,"related-org-nvidia-cuopt-routing-api-python":905},{"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},"cuopt-routing-api-python","solve vehicle routing problems with cuOpt","Vehicle routing (VRP, TSP, PDP) with cuOpt — Python API only. Use when the user is building or solving routing in Python.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,20],{"name":13,"slug":14,"type":15},"Routing","routing","tag",{"name":17,"slug":18,"type":15},"Optimization","optimization",{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},"Python","python",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-14T05:29:01.696483","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\u002Fcuopt-routing-api-python","---\nname: cuopt-routing-api-python\nversion: \"26.08.00\"\ndescription: Vehicle routing (VRP, TSP, PDP) with cuOpt — Python API only. Use when the user is building or solving routing in Python.\nlicense: Apache-2.0\nmetadata:\n  author: NVIDIA cuOpt Team\n  tags:\n    - cuopt\n    - routing\n    - vrp\n    - tsp\n    - python\n---\n\n\n\n# cuOpt Routing — Python API\n\nThis skill is **Python only**. Routing has no C API in cuOpt.\n\n## Required questions\n\nAsk these if not already clear:\n\n1. **Problem type** — TSP, VRP, or PDP?\n2. **Locations** — How many? Depot(s)? Cost or distance between pairs (matrix or derived)?\n3. **Orders \u002F tasks** — Which locations must be visited? Demand or service per stop?\n4. **Fleet** — Number of vehicles, capacity per vehicle (and per dimension if multiple), start\u002Fend locations?\n5. **Constraints** — Time windows (earliest\u002Flatest arrival), service times, precedence (order A before B)?\n\n## Minimal VRP Example\n\n```python\nimport cudf\nfrom cuopt import routing\n\ncost_matrix = cudf.DataFrame([...], dtype=\"float32\")\ndm = routing.DataModel(n_locations=4, n_fleet=2, n_orders=3)\ndm.add_cost_matrix(cost_matrix)\ndm.set_order_locations(cudf.Series([1, 2, 3], dtype=\"int32\"))\nsolution = routing.Solve(dm, routing.SolverSettings())\n\nif solution.get_status() == 0:\n    solution.display_routes()\n```\n\n## Adding Constraints\n\n```python\n# Time windows\ndm.add_transit_time_matrix(transit_time_matrix)\ndm.set_order_time_windows(earliest_series, latest_series)\n\n# Capacities\ndm.add_capacity_dimension(\"weight\", demand_series, capacity_series)\ndm.set_order_service_times(service_times)\ndm.set_vehicle_locations(start_locations, end_locations)\ndm.set_vehicle_time_windows(earliest_start, latest_return)\n\n# Pickup-delivery pairs\ndm.set_pickup_delivery_pairs(pickup_indices, delivery_indices)\n\n# Precedence\ndm.add_order_precedence(node_id=2, preceding_nodes=np.array([0, 1]))\n```\n\n## Solution Checking\n\n```python\nstatus = solution.get_status()  # 0=SUCCESS, 1=FAIL, 2=TIMEOUT, 3=EMPTY\nif status == 0:\n    route_df = solution.get_route()\n    total_cost = solution.get_total_objective()\nelse:\n    print(solution.get_error_message())\n    print(solution.get_infeasible_orders().to_list())\n```\n\n## Data Types (use explicit dtypes)\n\n```python\ncost_matrix = cost_matrix.astype(\"float32\")\norder_locations = cudf.Series([...], dtype=\"int32\")\ndemand = cudf.Series([...], dtype=\"int32\")\n```\n\n## Solver Settings\n\n```python\nss = routing.SolverSettings()\nss.set_time_limit(30)\nss.set_verbose_mode(True)\nss.set_error_logging_mode(True)\n```\n\n## Common Issues\n\n| Problem | Fix |\n|---------|-----|\n| Empty solution | Widen time windows or check travel times |\n| Infeasible orders | Increase fleet or capacity |\n| Status != 0 with time windows | Add `add_transit_time_matrix()` |\n| Wrong cost | Check cost_matrix is symmetric |\n| `compute_waypoint_sequence` alters route_df | It replaces the `location` column with waypoint ids in place — pass `route_df.copy()` if you still need cost-matrix indices (e.g. when iterating per truck) |\n\n## Debugging\n\n**When status != 0:** `print(solution.get_error_message())` and `print(solution.get_infeasible_orders().to_list())` to see which orders are infeasible.\n\n**Data types:** Use explicit dtypes (float32, int32) for matrices and series to avoid silent errors.\n\n## Examples\n\n- [examples.md](references\u002Fexamples.md) — VRP, PDP, multi-depot\n- [server_examples.md](references\u002Fserver_examples.md) — REST client (curl, Python)\n- **Reference models:** This skill's `assets\u002F` — [vrp_basic](assets\u002Fvrp_basic\u002F), [pdp_basic](assets\u002Fpdp_basic\u002F). See [assets\u002FREADME.md](assets\u002FREADME.md).\n\n## Escalate\n\nFor contribution or build-from-source, see the developer skill.\n",{"data":35,"body":43},{"name":4,"version":36,"description":6,"license":26,"metadata":37},"26.08.00",{"author":38,"tags":39},"NVIDIA cuOpt Team",[40,14,41,42,22],"cuopt","vrp","tsp",{"type":44,"children":45},"root",[46,55,69,76,81,136,142,254,260,388,394,457,463,494,500,539,545,666,672,698,708,714,782,788,793],{"type":47,"tag":48,"props":49,"children":51},"element","h1",{"id":50},"cuopt-routing-python-api",[52],{"type":53,"value":54},"text","cuOpt Routing — Python API",{"type":47,"tag":56,"props":57,"children":58},"p",{},[59,61,67],{"type":53,"value":60},"This skill is ",{"type":47,"tag":62,"props":63,"children":64},"strong",{},[65],{"type":53,"value":66},"Python only",{"type":53,"value":68},". Routing has no C API in cuOpt.",{"type":47,"tag":70,"props":71,"children":73},"h2",{"id":72},"required-questions",[74],{"type":53,"value":75},"Required questions",{"type":47,"tag":56,"props":77,"children":78},{},[79],{"type":53,"value":80},"Ask these if not already clear:",{"type":47,"tag":82,"props":83,"children":84},"ol",{},[85,96,106,116,126],{"type":47,"tag":86,"props":87,"children":88},"li",{},[89,94],{"type":47,"tag":62,"props":90,"children":91},{},[92],{"type":53,"value":93},"Problem type",{"type":53,"value":95}," — TSP, VRP, or PDP?",{"type":47,"tag":86,"props":97,"children":98},{},[99,104],{"type":47,"tag":62,"props":100,"children":101},{},[102],{"type":53,"value":103},"Locations",{"type":53,"value":105}," — How many? Depot(s)? Cost or distance between pairs (matrix or derived)?",{"type":47,"tag":86,"props":107,"children":108},{},[109,114],{"type":47,"tag":62,"props":110,"children":111},{},[112],{"type":53,"value":113},"Orders \u002F tasks",{"type":53,"value":115}," — Which locations must be visited? Demand or service per stop?",{"type":47,"tag":86,"props":117,"children":118},{},[119,124],{"type":47,"tag":62,"props":120,"children":121},{},[122],{"type":53,"value":123},"Fleet",{"type":53,"value":125}," — Number of vehicles, capacity per vehicle (and per dimension if multiple), start\u002Fend locations?",{"type":47,"tag":86,"props":127,"children":128},{},[129,134],{"type":47,"tag":62,"props":130,"children":131},{},[132],{"type":53,"value":133},"Constraints",{"type":53,"value":135}," — Time windows (earliest\u002Flatest arrival), service times, precedence (order A before B)?",{"type":47,"tag":70,"props":137,"children":139},{"id":138},"minimal-vrp-example",[140],{"type":53,"value":141},"Minimal VRP Example",{"type":47,"tag":143,"props":144,"children":148},"pre",{"className":145,"code":146,"language":22,"meta":147,"style":147},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","import cudf\nfrom cuopt import routing\n\ncost_matrix = cudf.DataFrame([...], dtype=\"float32\")\ndm = routing.DataModel(n_locations=4, n_fleet=2, n_orders=3)\ndm.add_cost_matrix(cost_matrix)\ndm.set_order_locations(cudf.Series([1, 2, 3], dtype=\"int32\"))\nsolution = routing.Solve(dm, routing.SolverSettings())\n\nif solution.get_status() == 0:\n    solution.display_routes()\n","",[149],{"type":47,"tag":150,"props":151,"children":152},"code",{"__ignoreMap":147},[153,164,173,183,192,201,210,219,228,236,245],{"type":47,"tag":154,"props":155,"children":158},"span",{"class":156,"line":157},"line",1,[159],{"type":47,"tag":154,"props":160,"children":161},{},[162],{"type":53,"value":163},"import cudf\n",{"type":47,"tag":154,"props":165,"children":167},{"class":156,"line":166},2,[168],{"type":47,"tag":154,"props":169,"children":170},{},[171],{"type":53,"value":172},"from cuopt import routing\n",{"type":47,"tag":154,"props":174,"children":176},{"class":156,"line":175},3,[177],{"type":47,"tag":154,"props":178,"children":180},{"emptyLinePlaceholder":179},true,[181],{"type":53,"value":182},"\n",{"type":47,"tag":154,"props":184,"children":186},{"class":156,"line":185},4,[187],{"type":47,"tag":154,"props":188,"children":189},{},[190],{"type":53,"value":191},"cost_matrix = cudf.DataFrame([...], dtype=\"float32\")\n",{"type":47,"tag":154,"props":193,"children":195},{"class":156,"line":194},5,[196],{"type":47,"tag":154,"props":197,"children":198},{},[199],{"type":53,"value":200},"dm = routing.DataModel(n_locations=4, n_fleet=2, n_orders=3)\n",{"type":47,"tag":154,"props":202,"children":204},{"class":156,"line":203},6,[205],{"type":47,"tag":154,"props":206,"children":207},{},[208],{"type":53,"value":209},"dm.add_cost_matrix(cost_matrix)\n",{"type":47,"tag":154,"props":211,"children":213},{"class":156,"line":212},7,[214],{"type":47,"tag":154,"props":215,"children":216},{},[217],{"type":53,"value":218},"dm.set_order_locations(cudf.Series([1, 2, 3], dtype=\"int32\"))\n",{"type":47,"tag":154,"props":220,"children":222},{"class":156,"line":221},8,[223],{"type":47,"tag":154,"props":224,"children":225},{},[226],{"type":53,"value":227},"solution = routing.Solve(dm, routing.SolverSettings())\n",{"type":47,"tag":154,"props":229,"children":231},{"class":156,"line":230},9,[232],{"type":47,"tag":154,"props":233,"children":234},{"emptyLinePlaceholder":179},[235],{"type":53,"value":182},{"type":47,"tag":154,"props":237,"children":239},{"class":156,"line":238},10,[240],{"type":47,"tag":154,"props":241,"children":242},{},[243],{"type":53,"value":244},"if solution.get_status() == 0:\n",{"type":47,"tag":154,"props":246,"children":248},{"class":156,"line":247},11,[249],{"type":47,"tag":154,"props":250,"children":251},{},[252],{"type":53,"value":253},"    solution.display_routes()\n",{"type":47,"tag":70,"props":255,"children":257},{"id":256},"adding-constraints",[258],{"type":53,"value":259},"Adding Constraints",{"type":47,"tag":143,"props":261,"children":263},{"className":145,"code":262,"language":22,"meta":147,"style":147},"# Time windows\ndm.add_transit_time_matrix(transit_time_matrix)\ndm.set_order_time_windows(earliest_series, latest_series)\n\n# Capacities\ndm.add_capacity_dimension(\"weight\", demand_series, capacity_series)\ndm.set_order_service_times(service_times)\ndm.set_vehicle_locations(start_locations, end_locations)\ndm.set_vehicle_time_windows(earliest_start, latest_return)\n\n# Pickup-delivery pairs\ndm.set_pickup_delivery_pairs(pickup_indices, delivery_indices)\n\n# Precedence\ndm.add_order_precedence(node_id=2, preceding_nodes=np.array([0, 1]))\n",[264],{"type":47,"tag":150,"props":265,"children":266},{"__ignoreMap":147},[267,275,283,291,298,306,314,322,330,338,345,353,362,370,379],{"type":47,"tag":154,"props":268,"children":269},{"class":156,"line":157},[270],{"type":47,"tag":154,"props":271,"children":272},{},[273],{"type":53,"value":274},"# Time windows\n",{"type":47,"tag":154,"props":276,"children":277},{"class":156,"line":166},[278],{"type":47,"tag":154,"props":279,"children":280},{},[281],{"type":53,"value":282},"dm.add_transit_time_matrix(transit_time_matrix)\n",{"type":47,"tag":154,"props":284,"children":285},{"class":156,"line":175},[286],{"type":47,"tag":154,"props":287,"children":288},{},[289],{"type":53,"value":290},"dm.set_order_time_windows(earliest_series, latest_series)\n",{"type":47,"tag":154,"props":292,"children":293},{"class":156,"line":185},[294],{"type":47,"tag":154,"props":295,"children":296},{"emptyLinePlaceholder":179},[297],{"type":53,"value":182},{"type":47,"tag":154,"props":299,"children":300},{"class":156,"line":194},[301],{"type":47,"tag":154,"props":302,"children":303},{},[304],{"type":53,"value":305},"# Capacities\n",{"type":47,"tag":154,"props":307,"children":308},{"class":156,"line":203},[309],{"type":47,"tag":154,"props":310,"children":311},{},[312],{"type":53,"value":313},"dm.add_capacity_dimension(\"weight\", demand_series, capacity_series)\n",{"type":47,"tag":154,"props":315,"children":316},{"class":156,"line":212},[317],{"type":47,"tag":154,"props":318,"children":319},{},[320],{"type":53,"value":321},"dm.set_order_service_times(service_times)\n",{"type":47,"tag":154,"props":323,"children":324},{"class":156,"line":221},[325],{"type":47,"tag":154,"props":326,"children":327},{},[328],{"type":53,"value":329},"dm.set_vehicle_locations(start_locations, end_locations)\n",{"type":47,"tag":154,"props":331,"children":332},{"class":156,"line":230},[333],{"type":47,"tag":154,"props":334,"children":335},{},[336],{"type":53,"value":337},"dm.set_vehicle_time_windows(earliest_start, latest_return)\n",{"type":47,"tag":154,"props":339,"children":340},{"class":156,"line":238},[341],{"type":47,"tag":154,"props":342,"children":343},{"emptyLinePlaceholder":179},[344],{"type":53,"value":182},{"type":47,"tag":154,"props":346,"children":347},{"class":156,"line":247},[348],{"type":47,"tag":154,"props":349,"children":350},{},[351],{"type":53,"value":352},"# Pickup-delivery pairs\n",{"type":47,"tag":154,"props":354,"children":356},{"class":156,"line":355},12,[357],{"type":47,"tag":154,"props":358,"children":359},{},[360],{"type":53,"value":361},"dm.set_pickup_delivery_pairs(pickup_indices, delivery_indices)\n",{"type":47,"tag":154,"props":363,"children":365},{"class":156,"line":364},13,[366],{"type":47,"tag":154,"props":367,"children":368},{"emptyLinePlaceholder":179},[369],{"type":53,"value":182},{"type":47,"tag":154,"props":371,"children":373},{"class":156,"line":372},14,[374],{"type":47,"tag":154,"props":375,"children":376},{},[377],{"type":53,"value":378},"# Precedence\n",{"type":47,"tag":154,"props":380,"children":382},{"class":156,"line":381},15,[383],{"type":47,"tag":154,"props":384,"children":385},{},[386],{"type":53,"value":387},"dm.add_order_precedence(node_id=2, preceding_nodes=np.array([0, 1]))\n",{"type":47,"tag":70,"props":389,"children":391},{"id":390},"solution-checking",[392],{"type":53,"value":393},"Solution Checking",{"type":47,"tag":143,"props":395,"children":397},{"className":145,"code":396,"language":22,"meta":147,"style":147},"status = solution.get_status()  # 0=SUCCESS, 1=FAIL, 2=TIMEOUT, 3=EMPTY\nif status == 0:\n    route_df = solution.get_route()\n    total_cost = solution.get_total_objective()\nelse:\n    print(solution.get_error_message())\n    print(solution.get_infeasible_orders().to_list())\n",[398],{"type":47,"tag":150,"props":399,"children":400},{"__ignoreMap":147},[401,409,417,425,433,441,449],{"type":47,"tag":154,"props":402,"children":403},{"class":156,"line":157},[404],{"type":47,"tag":154,"props":405,"children":406},{},[407],{"type":53,"value":408},"status = solution.get_status()  # 0=SUCCESS, 1=FAIL, 2=TIMEOUT, 3=EMPTY\n",{"type":47,"tag":154,"props":410,"children":411},{"class":156,"line":166},[412],{"type":47,"tag":154,"props":413,"children":414},{},[415],{"type":53,"value":416},"if status == 0:\n",{"type":47,"tag":154,"props":418,"children":419},{"class":156,"line":175},[420],{"type":47,"tag":154,"props":421,"children":422},{},[423],{"type":53,"value":424},"    route_df = solution.get_route()\n",{"type":47,"tag":154,"props":426,"children":427},{"class":156,"line":185},[428],{"type":47,"tag":154,"props":429,"children":430},{},[431],{"type":53,"value":432},"    total_cost = solution.get_total_objective()\n",{"type":47,"tag":154,"props":434,"children":435},{"class":156,"line":194},[436],{"type":47,"tag":154,"props":437,"children":438},{},[439],{"type":53,"value":440},"else:\n",{"type":47,"tag":154,"props":442,"children":443},{"class":156,"line":203},[444],{"type":47,"tag":154,"props":445,"children":446},{},[447],{"type":53,"value":448},"    print(solution.get_error_message())\n",{"type":47,"tag":154,"props":450,"children":451},{"class":156,"line":212},[452],{"type":47,"tag":154,"props":453,"children":454},{},[455],{"type":53,"value":456},"    print(solution.get_infeasible_orders().to_list())\n",{"type":47,"tag":70,"props":458,"children":460},{"id":459},"data-types-use-explicit-dtypes",[461],{"type":53,"value":462},"Data Types (use explicit dtypes)",{"type":47,"tag":143,"props":464,"children":466},{"className":145,"code":465,"language":22,"meta":147,"style":147},"cost_matrix = cost_matrix.astype(\"float32\")\norder_locations = cudf.Series([...], dtype=\"int32\")\ndemand = cudf.Series([...], dtype=\"int32\")\n",[467],{"type":47,"tag":150,"props":468,"children":469},{"__ignoreMap":147},[470,478,486],{"type":47,"tag":154,"props":471,"children":472},{"class":156,"line":157},[473],{"type":47,"tag":154,"props":474,"children":475},{},[476],{"type":53,"value":477},"cost_matrix = cost_matrix.astype(\"float32\")\n",{"type":47,"tag":154,"props":479,"children":480},{"class":156,"line":166},[481],{"type":47,"tag":154,"props":482,"children":483},{},[484],{"type":53,"value":485},"order_locations = cudf.Series([...], dtype=\"int32\")\n",{"type":47,"tag":154,"props":487,"children":488},{"class":156,"line":175},[489],{"type":47,"tag":154,"props":490,"children":491},{},[492],{"type":53,"value":493},"demand = cudf.Series([...], dtype=\"int32\")\n",{"type":47,"tag":70,"props":495,"children":497},{"id":496},"solver-settings",[498],{"type":53,"value":499},"Solver Settings",{"type":47,"tag":143,"props":501,"children":503},{"className":145,"code":502,"language":22,"meta":147,"style":147},"ss = routing.SolverSettings()\nss.set_time_limit(30)\nss.set_verbose_mode(True)\nss.set_error_logging_mode(True)\n",[504],{"type":47,"tag":150,"props":505,"children":506},{"__ignoreMap":147},[507,515,523,531],{"type":47,"tag":154,"props":508,"children":509},{"class":156,"line":157},[510],{"type":47,"tag":154,"props":511,"children":512},{},[513],{"type":53,"value":514},"ss = routing.SolverSettings()\n",{"type":47,"tag":154,"props":516,"children":517},{"class":156,"line":166},[518],{"type":47,"tag":154,"props":519,"children":520},{},[521],{"type":53,"value":522},"ss.set_time_limit(30)\n",{"type":47,"tag":154,"props":524,"children":525},{"class":156,"line":175},[526],{"type":47,"tag":154,"props":527,"children":528},{},[529],{"type":53,"value":530},"ss.set_verbose_mode(True)\n",{"type":47,"tag":154,"props":532,"children":533},{"class":156,"line":185},[534],{"type":47,"tag":154,"props":535,"children":536},{},[537],{"type":53,"value":538},"ss.set_error_logging_mode(True)\n",{"type":47,"tag":70,"props":540,"children":542},{"id":541},"common-issues",[543],{"type":53,"value":544},"Common Issues",{"type":47,"tag":546,"props":547,"children":548},"table",{},[549,568],{"type":47,"tag":550,"props":551,"children":552},"thead",{},[553],{"type":47,"tag":554,"props":555,"children":556},"tr",{},[557,563],{"type":47,"tag":558,"props":559,"children":560},"th",{},[561],{"type":53,"value":562},"Problem",{"type":47,"tag":558,"props":564,"children":565},{},[566],{"type":53,"value":567},"Fix",{"type":47,"tag":569,"props":570,"children":571},"tbody",{},[572,586,599,618,631],{"type":47,"tag":554,"props":573,"children":574},{},[575,581],{"type":47,"tag":576,"props":577,"children":578},"td",{},[579],{"type":53,"value":580},"Empty solution",{"type":47,"tag":576,"props":582,"children":583},{},[584],{"type":53,"value":585},"Widen time windows or check travel times",{"type":47,"tag":554,"props":587,"children":588},{},[589,594],{"type":47,"tag":576,"props":590,"children":591},{},[592],{"type":53,"value":593},"Infeasible orders",{"type":47,"tag":576,"props":595,"children":596},{},[597],{"type":53,"value":598},"Increase fleet or capacity",{"type":47,"tag":554,"props":600,"children":601},{},[602,607],{"type":47,"tag":576,"props":603,"children":604},{},[605],{"type":53,"value":606},"Status != 0 with time windows",{"type":47,"tag":576,"props":608,"children":609},{},[610,612],{"type":53,"value":611},"Add ",{"type":47,"tag":150,"props":613,"children":615},{"className":614},[],[616],{"type":53,"value":617},"add_transit_time_matrix()",{"type":47,"tag":554,"props":619,"children":620},{},[621,626],{"type":47,"tag":576,"props":622,"children":623},{},[624],{"type":53,"value":625},"Wrong cost",{"type":47,"tag":576,"props":627,"children":628},{},[629],{"type":53,"value":630},"Check cost_matrix is symmetric",{"type":47,"tag":554,"props":632,"children":633},{},[634,645],{"type":47,"tag":576,"props":635,"children":636},{},[637,643],{"type":47,"tag":150,"props":638,"children":640},{"className":639},[],[641],{"type":53,"value":642},"compute_waypoint_sequence",{"type":53,"value":644}," alters route_df",{"type":47,"tag":576,"props":646,"children":647},{},[648,650,656,658,664],{"type":53,"value":649},"It replaces the ",{"type":47,"tag":150,"props":651,"children":653},{"className":652},[],[654],{"type":53,"value":655},"location",{"type":53,"value":657}," column with waypoint ids in place — pass ",{"type":47,"tag":150,"props":659,"children":661},{"className":660},[],[662],{"type":53,"value":663},"route_df.copy()",{"type":53,"value":665}," if you still need cost-matrix indices (e.g. when iterating per truck)",{"type":47,"tag":70,"props":667,"children":669},{"id":668},"debugging",[670],{"type":53,"value":671},"Debugging",{"type":47,"tag":56,"props":673,"children":674},{},[675,680,682,688,690,696],{"type":47,"tag":62,"props":676,"children":677},{},[678],{"type":53,"value":679},"When status != 0:",{"type":53,"value":681}," ",{"type":47,"tag":150,"props":683,"children":685},{"className":684},[],[686],{"type":53,"value":687},"print(solution.get_error_message())",{"type":53,"value":689}," and ",{"type":47,"tag":150,"props":691,"children":693},{"className":692},[],[694],{"type":53,"value":695},"print(solution.get_infeasible_orders().to_list())",{"type":53,"value":697}," to see which orders are infeasible.",{"type":47,"tag":56,"props":699,"children":700},{},[701,706],{"type":47,"tag":62,"props":702,"children":703},{},[704],{"type":53,"value":705},"Data types:",{"type":53,"value":707}," Use explicit dtypes (float32, int32) for matrices and series to avoid silent errors.",{"type":47,"tag":70,"props":709,"children":711},{"id":710},"examples",[712],{"type":53,"value":713},"Examples",{"type":47,"tag":715,"props":716,"children":717},"ul",{},[718,730,741],{"type":47,"tag":86,"props":719,"children":720},{},[721,728],{"type":47,"tag":722,"props":723,"children":725},"a",{"href":724},"references\u002Fexamples.md",[726],{"type":53,"value":727},"examples.md",{"type":53,"value":729}," — VRP, PDP, multi-depot",{"type":47,"tag":86,"props":731,"children":732},{},[733,739],{"type":47,"tag":722,"props":734,"children":736},{"href":735},"references\u002Fserver_examples.md",[737],{"type":53,"value":738},"server_examples.md",{"type":53,"value":740}," — REST client (curl, Python)",{"type":47,"tag":86,"props":742,"children":743},{},[744,749,751,757,759,765,767,773,775,780],{"type":47,"tag":62,"props":745,"children":746},{},[747],{"type":53,"value":748},"Reference models:",{"type":53,"value":750}," This skill's ",{"type":47,"tag":150,"props":752,"children":754},{"className":753},[],[755],{"type":53,"value":756},"assets\u002F",{"type":53,"value":758}," — ",{"type":47,"tag":722,"props":760,"children":762},{"href":761},"assets\u002Fvrp_basic\u002F",[763],{"type":53,"value":764},"vrp_basic",{"type":53,"value":766},", ",{"type":47,"tag":722,"props":768,"children":770},{"href":769},"assets\u002Fpdp_basic\u002F",[771],{"type":53,"value":772},"pdp_basic",{"type":53,"value":774},". See ",{"type":47,"tag":722,"props":776,"children":778},{"href":777},"assets\u002FREADME.md",[779],{"type":53,"value":777},{"type":53,"value":781},".",{"type":47,"tag":70,"props":783,"children":785},{"id":784},"escalate",[786],{"type":53,"value":787},"Escalate",{"type":47,"tag":56,"props":789,"children":790},{},[791],{"type":53,"value":792},"For contribution or build-from-source, see the developer skill.",{"type":47,"tag":794,"props":795,"children":796},"style",{},[797],{"type":53,"value":798},"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":800,"total":904},[801,818,832,846,858,875,890],{"slug":802,"name":802,"fn":803,"description":804,"org":805,"tags":806,"stars":23,"repoUrl":24,"updatedAt":817},"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},[807,810,813,814],{"name":808,"slug":809,"type":15},"Data Analysis","data-analysis",{"name":811,"slug":812,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":815,"slug":816,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":819,"name":819,"fn":820,"description":821,"org":822,"tags":823,"stars":23,"repoUrl":24,"updatedAt":831},"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},[824,827,830],{"name":825,"slug":826,"type":15},"Deployment","deployment",{"name":828,"slug":829,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":833,"name":833,"fn":834,"description":835,"org":836,"tags":837,"stars":23,"repoUrl":24,"updatedAt":845},"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},[838,841,842],{"name":839,"slug":840,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":843,"slug":844,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":847,"name":847,"fn":848,"description":849,"org":850,"tags":851,"stars":23,"repoUrl":24,"updatedAt":857},"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},[852,853,854],{"name":808,"slug":809,"type":15},{"name":9,"slug":8,"type":15},{"name":855,"slug":856,"type":15},"Testing","testing","2026-07-17T05:29:03.913266",{"slug":859,"name":859,"fn":860,"description":861,"org":862,"tags":863,"stars":23,"repoUrl":24,"updatedAt":874},"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},[864,867,870,871],{"name":865,"slug":866,"type":15},"Automation","automation",{"name":868,"slug":869,"type":15},"Imaging","imaging",{"name":9,"slug":8,"type":15},{"name":872,"slug":873,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":876,"name":876,"fn":877,"description":878,"org":879,"tags":880,"stars":23,"repoUrl":24,"updatedAt":889},"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},[881,882,885,886],{"name":825,"slug":826,"type":15},{"name":883,"slug":884,"type":15},"Docker","docker",{"name":9,"slug":8,"type":15},{"name":887,"slug":888,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":891,"name":891,"fn":892,"description":893,"org":894,"tags":895,"stars":23,"repoUrl":24,"updatedAt":903},"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},[896,897,900],{"name":9,"slug":8,"type":15},{"name":898,"slug":899,"type":15},"Quantum Computing","quantum-computing",{"name":901,"slug":902,"type":15},"Simulation","simulation","2026-07-14T05:26:58.898253",305,{"items":906,"total":1053},[907,925,939,950,962,974,987,1001,1012,1021,1035,1044],{"slug":908,"name":908,"fn":909,"description":910,"org":911,"tags":912,"stars":922,"repoUrl":923,"updatedAt":924},"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},[913,916,919],{"name":914,"slug":915,"type":15},"Documentation","documentation",{"name":917,"slug":918,"type":15},"MCP","mcp",{"name":920,"slug":921,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":926,"name":926,"fn":927,"description":928,"org":929,"tags":930,"stars":936,"repoUrl":937,"updatedAt":938},"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},[931,934,935],{"name":932,"slug":933,"type":15},"Containers","containers",{"name":825,"slug":826,"type":15},{"name":21,"slug":22,"type":15},17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":940,"name":940,"fn":941,"description":942,"org":943,"tags":944,"stars":936,"repoUrl":937,"updatedAt":949},"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},[945,948],{"name":946,"slug":947,"type":15},"CI\u002FCD","ci-cd",{"name":825,"slug":826,"type":15},"2026-07-14T05:25:59.97109",{"slug":951,"name":951,"fn":952,"description":953,"org":954,"tags":955,"stars":936,"repoUrl":937,"updatedAt":961},"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},[956,957,958],{"name":946,"slug":947,"type":15},{"name":825,"slug":826,"type":15},{"name":959,"slug":960,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":963,"name":963,"fn":964,"description":965,"org":966,"tags":967,"stars":936,"repoUrl":937,"updatedAt":973},"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},[968,969,970],{"name":671,"slug":668,"type":15},{"name":959,"slug":960,"type":15},{"name":971,"slug":972,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":975,"name":975,"fn":976,"description":977,"org":978,"tags":979,"stars":936,"repoUrl":937,"updatedAt":986},"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},[980,983],{"name":981,"slug":982,"type":15},"Best Practices","best-practices",{"name":984,"slug":985,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":988,"name":988,"fn":989,"description":990,"org":991,"tags":992,"stars":936,"repoUrl":937,"updatedAt":1000},"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},[993,996,999],{"name":994,"slug":995,"type":15},"Machine Learning","machine-learning",{"name":997,"slug":998,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1002,"name":1002,"fn":1003,"description":1004,"org":1005,"tags":1006,"stars":936,"repoUrl":937,"updatedAt":1011},"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},[1007,1010],{"name":1008,"slug":1009,"type":15},"QA","qa",{"name":855,"slug":856,"type":15},"2026-07-14T05:25:53.673039",{"slug":1013,"name":1013,"fn":1014,"description":1015,"org":1016,"tags":1017,"stars":936,"repoUrl":937,"updatedAt":1020},"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},[1018,1019],{"name":825,"slug":826,"type":15},{"name":828,"slug":829,"type":15},"2026-07-14T05:25:49.362534",{"slug":1022,"name":1022,"fn":1023,"description":1024,"org":1025,"tags":1026,"stars":936,"repoUrl":937,"updatedAt":1034},"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},[1027,1030,1031],{"name":1028,"slug":1029,"type":15},"Code Review","code-review",{"name":959,"slug":960,"type":15},{"name":1032,"slug":1033,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1036,"name":1036,"fn":1037,"description":1038,"org":1039,"tags":1040,"stars":936,"repoUrl":937,"updatedAt":1043},"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},[1041,1042],{"name":1008,"slug":1009,"type":15},{"name":855,"slug":856,"type":15},"2026-07-14T05:25:54.928983",{"slug":1045,"name":1045,"fn":1046,"description":1047,"org":1048,"tags":1049,"stars":936,"repoUrl":937,"updatedAt":1052},"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},[1050,1051],{"name":865,"slug":866,"type":15},{"name":946,"slug":947,"type":15},"2026-07-30T05:29:03.275638",496]