{"schema_version":"onlylabs.public_analysis_evidence.v1","title":"NVIDIA analysis evidence pack","description":"Public onlylabs evidence pack for cited agent analysis: captured pages, ranked public signals, and stored web-search provenance used by the background analysis workflow.","url":"https://onlylabs.fyi/analysis/nvidia","json_url":"https://onlylabs.fyi/analysis/nvidia/evidence.json","generated_at":"2026-06-11T15:10:45.571Z","org":{"slug":"nvidia","name":"NVIDIA","category":"frontier-lab","category_label":"Frontier lab","dossier_url":"https://onlylabs.fyi/labs/nvidia"},"analysis":{"url":"https://onlylabs.fyi/analysis/nvidia","json_url":"https://onlylabs.fyi/analysis/nvidia/analysis.json","generated_at":"2026-06-08T15:59:09.594+00:00"},"workflow":{"version":"synthesize-analyses","provider":null,"model":null,"agent":null,"public_pack_mode":"local-pages-and-events","live_web_fetches":false,"note":"Public evidence exports do not trigger live Exa calls; stored Exa provenance is included when analysis metadata contains it."},"stats":{"pages":28,"events":140,"web":0,"evidence":88,"signal_desks":{"hiring":0,"forks":0,"releases":48,"talking":12,"repos":0},"data_radar_lanes":{"data":2,"evals":1,"infrastructure":8,"safety":1,"product":1},"data_radar_matches":9,"stored_analysis_evidence":null,"stored_analysis_web":null,"stored_analysis_signal_desks":null,"stored_analysis_data_radar_lanes":null,"stored_analysis_data_radar_matches":null},"stored_web_provenance":null,"evidence":[{"ref":"P1","kind":"page","title":"NVIDIA/cuDecomp repository metadata","date":"2026-06-11T07:04:14.146816+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/cuDecomp","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/cuDecomp\n\nDescription: An Adaptive Pencil Decomposition Library for NVIDIA GPUs\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 87\n\nForks: 15\n\nOpen issues: 1\n\nCreated: 2022-06-21T15:52:42Z\n\nPushed: 2026-06-10T20:33:08Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# cuDecomp\n\nAn Adaptive Pencil Decomposition Library for NVIDIA GPUs\n\n## Introduction\n\ncuDecomp is a library for managing 1D (slab) and 2D (pencil) parallel decompositions of 3D Cartesian spatial domains on NVIDIA GPUs, with routines to perform global transpositions and halo communications. The library is inspired by the [2DECOMP&FFT Fortran library](https://github.com/xcompact3d/2decomp-fft), a popular decomposition library for numerical simulation codes, with a similar set of available transposition routines. While 2DECOMP&FFT and similar libraries in the past have been written to target CPU systems, this library is designed for GPU systems, leveraging CUDA-aware MPI and additional communication libraries optimized for GPUs, like the [NVIDIA Collective Communication Library (NCCL)](https://github.com/NVIDIA/nccl) and [NVIDIA OpenSHMEM Library (NVSHMEM)](https://developer.nvidia.com/nvshmem).\n\nPlease refer to the [documentation](https://nvidia.github.io/cuDecomp/) for additional information on the library and usage details.\n\nThis library is currently in a research-oriented state, and has been released as a companion to a paper presented at the PASC22 conference ([link](https://dl.acm.org/doi/10.1145/3539781.3539797)). We are making it available here as it can be useful in other applications outside of this study or as a benchmarking tool and usage example for various GPU communication libraries to perform transpose and halo communication.\n\nPlease contact us or open a GitHub issue if you are interested in using this library in your own solvers and have questions on usage and/or feature requests.\n\n## Build\nYou can build this library using CMake. A CMake build of the library without additional examples/tests can be completed using the following commands\n```shell\n$ mkdir build\n$ cd build\n$ cmake ..\n$ make -j\n```\nThere are several build variables available to configure the CMake build which c"},{"ref":"P2","kind":"page","title":"NVIDIA/TorchFort repository metadata","date":"2026-06-11T07:04:13.487763+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/TorchFort","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/TorchFort\n\nDescription: An Online Deep Learning Interface for HPC programs on NVIDIA GPUs\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 196\n\nForks: 35\n\nOpen issues: 3\n\nCreated: 2023-08-01T18:40:09Z\n\nPushed: 2026-06-11T00:34:11Z\n\nDefault branch: master\n\nFork: no\n\nArchived: no\n\nREADME:\n# TorchFort\n\nAn Online Deep Learning Interface for HPC programs on NVIDIA GPUs\n\n## Introduction\nTorchFort is a DL training and inference interface for HPC programs implemented using LibTorch, the C++ backend used by the [PyTorch](https://pytorch.org) framework.\nThe goal of this library is to help practitioners and domain scientists to seamlessly combine their simulation codes with Deep Learning functionalities available \nwithin PyTorch.\nThis library can be invoked directly from Fortran or C/C++ programs, enabling transparent sharing of data arrays to and from the DL framework all contained within the\nsimulation process (i.e., no external glue/data-sharing code required). The library can directly load PyTorch model definitions exported to TorchScript and implements a\nconfigurable training process that users can control via a simple YAML configuration file format. The configuration files enable users to specify optimizer and loss selection,\nlearning rate schedules, and much more.\n\nPlease refer to the [documentation](https://nvidia.github.io/TorchFort/) for additional information on the library, build instructions, and usage details.\n\nPlease refer to the [examples](examples) to see TorchFort in action.\n\nContact us or open a GitHub issue if you are interested in using this library in your own solvers and have questions on usage and/or feature requests.\n\n## License\nThis library is released under an Apache 2.0 license, which can be found in [LICENSE](LICENSE)."},{"ref":"P3","kind":"page","title":"NVIDIA/bmcweb repository metadata","date":"2026-06-11T07:04:13.34759+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/bmcweb","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/bmcweb\n\nDescription: A do everything Redfish, KVM, GUI, and DBus webserver for OpenBMC\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 7\n\nForks: 7\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:46:17Z\n\nPushed: 2026-06-10T18:22:41Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/bmcweb\n\nArchived: no\n\nREADME:\n# OpenBMC webserver\n\nThis component attempts to be a \"do everything\" embedded webserver for OpenBMC.\n\n## Features\n\nThe webserver implements a few distinct interfaces:\n\n- DBus event websocket. Allows registering for changes to specific dbus paths,\nproperties, and will send an event from the websocket if those filters match.\n- OpenBMC DBus REST API. Allows direct, low interference, high fidelity access\nto dbus and the objects it represents.\n- Serial: A serial websocket for interacting with the host serial console\nthrough websockets.\n- Redfish: A protocol compliant, [DBus to Redfish translator](docs/Redfish.md).\n- KVM: A websocket based implementation of the RFB (VNC) frame buffer protocol\nintended to mate to webui-vue to provide a complete KVM implementation.\n\n## Protocols\n\nbmcweb at a protocol level supports http and https. TLS is supported through\nOpenSSL. HTTP/1 and HTTP/2 are supported using ALPN registration for TLS\nconnections and h2c upgrade header for http connections.\n\n## AuthX\n\n### Authentication\n\nBmcweb supports multiple authentication protocols:\n\n- Basic authentication per RFC7617\n- Cookie based authentication for authenticating against webui-vue\n- Mutual TLS authentication based on OpenSSL\n- Session authentication through webui-vue\n- XToken based authentication conformant to Redfish DSP0266\n\nEach of these types of authentication is able to be enabled or disabled both via\nruntime policy changes (through the relevant Redfish APIs) or via configure time\noptions. All authentication mechanisms supporting username/password are routed\nto libpam, to allow for customization in authentication implementations.\n\n### Authorization\n\nAll authorization in bmcweb is determined at routing time, and per route, and\nconforms to the Redfish PrivilegeRegistry.\n\n\\*Note: Non-Redfish functions are mapped to the closest equivalent Redfish\nprivilege level.\n\n## C"},{"ref":"P4","kind":"page","title":"NVIDIA/dbus-sensors repository metadata","date":"2026-06-11T07:04:13.297118+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/dbus-sensors","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/dbus-sensors\n\nDescription: D-Bus configurable sensor scanning applications\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 6\n\nForks: 2\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:21Z\n\nPushed: 2026-06-10T18:22:34Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/dbus-sensors\n\nArchived: no\n\nREADME:\n# dbus-sensors\n\ndbus-sensors is a collection of sensor applications that provide the\nxyz.openbmc_project.Sensor collection of interfaces. They read sensor values\nfrom hwmon, d-bus, or direct driver access to provide readings. Some advance\nnon-sensor features such as fan presence, pwm control, and automatic cpu\ndetection (x86) are also supported.\n\n## key features\n\n- runtime re-configurable from d-bus (entity-manager or the like)\n\n- isolated: each sensor type is isolated into its own daemon, so a bug in one\nsensor is unlikely to affect another, and single sensor modifications are\npossible\n\n- async single-threaded: uses sdbusplus/asio bindings\n\n- multiple data inputs: hwmon, d-bus, direct driver access\n\n## dbus interfaces\n\nA typical dbus-sensors object support the following dbus interfaces:\n\n```text\nPath /xyz/openbmc_project/sensors/<type>/<sensor_name>\n\nInterfaces xyz.openbmc_project.Sensor.Value\nxyz.openbmc_project.Sensor.Threshold.Critical\nxyz.openbmc_project.Sensor.Threshold.Warning\nxyz.openbmc_project.State.Decorator.Availability\nxyz.openbmc_project.State.Decorator.OperationalStatus\nxyz.openbmc_project.Association.Definitions\n\n```\n\nSensor interfaces collection are described in\n[phosphor-dbus-interfaces](https://github.com/openbmc/phosphor-dbus-interfaces/tree/master/yaml/xyz/openbmc_project/Sensor).\n\nConsumer examples of these interfaces are\n[Redfish](https://github.com/openbmc/bmcweb/blob/master/redfish-core/lib/sensors.hpp),\n[Phosphor-Pid-Control](https://github.com/openbmc/phosphor-pid-control),\n[IPMI SDR](https://github.com/openbmc/phosphor-host-ipmid/blob/master/dbus-sdr/sensorcommands.cpp).\n\n## Reactor\n\ndbus-sensor daemons are [reactors](https://github.com/openbmc/entity-manager)\nthat dynamically create and update sensors configuration when system\nconfiguration gets updated.\n\nUsing asio timers and async calls, dbus-sensor daemons read sensor v"},{"ref":"P5","kind":"page","title":"NVIDIA/libpldm repository metadata","date":"2026-06-11T07:04:13.119475+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/libpldm","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/libpldm\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 2\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:28Z\n\nPushed: 2026-06-10T18:07:58Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/libpldm\n\nArchived: no\n\nREADME:\n# libpldm\n\nThis is a library which deals with the encoding and decoding of PLDM messages.\nIt should be possible to use this library by projects other than OpenBMC, and\nhence certain constraints apply to it:\n\n- keeping it light weight\n- implementation in C\n- minimal dynamic memory allocations\n- endian-safe\n- no OpenBMC specific dependencies\n\nSource files are named according to the PLDM Type, for eg base.[h/c], fru.[h/c],\netc.\n\nGiven a PLDM command \"foo\", the library will provide the following API: For the\nRequester function:\n\n```c\nencode_foo_req() - encode a foo request\ndecode_foo_resp() - decode a response to foo\n```\n\nFor the Responder function:\n\n```c\ndecode_foo_req() - decode a foo request\nencode_foo_resp() - encode a response to foo\n```\n\nThe library also provides API to pack and unpack PLDM headers.\n\n## To Build\n\n`libpldm` is configured and built using `meson`. Python's `pip` or\n[`pipx`][pipx] can be used to install a recent version on your machine:\n\n[pipx]: https://pipx.pypa.io/latest/\n\n```sh\npipx install meson\n```\n\nOnce `meson` is installed:\n\n```sh\nmeson setup build && meson compile -C build\n```\n\n## To run unit tests\n\n```sh\nmeson test -C build\n```\n\n## Working with `libpldm`\n\nComponents of the library ABI[^1] (loosely, functions) are separated into three\ncategories:\n\n[^1]: [\"library API + compiler ABI = library ABI\"][libstdc++-library-abi]\n\n[libstdc++-library-abi]:\nhttps://gcc.gnu.org/onlinedocs/libstdc++/manual/abi.html\n\n1. Stable\n2. Testing\n3. Deprecated\n\nApplications depending on `libpldm` should aim to only use functions from the\nstable category. However, this may not always be possible. What to do when\nrequired functions fall into the deprecated or testing categories is discussed\nin [CONTRIBUTING](CONTRIBUTING.md#Library-background).\n\n### Upgrading libpldm\n\nlibpldm is maintained with the expectation that users move between successive\nreleases when upgrading. This constraint allows the library to reintroduce types\n"},{"ref":"P6","kind":"page","title":"NVIDIA/TorchFort v0.1.0","date":"2026-06-11T07:04:13.09663+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/TorchFort/releases/tag/v0.1.0","signal_url":null,"signal_json_url":null,"text":"# v0.1.0\n\nRepository: NVIDIA/TorchFort\n\nTag: v0.1.0\n\nPublished: 2023-08-01T21:55:43Z\n\nPrerelease: yes\n\nRelease notes:\nInitial release of TorchFort."},{"ref":"P7","kind":"page","title":"NVIDIA/linux repository metadata","date":"2026-06-11T07:04:12.611398+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/linux","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/linux\n\nDescription: OpenBMC Linux kernel source tree\n\nLanguage: C\n\nLicense: NOASSERTION\n\nStars: 12\n\nForks: 7\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:30Z\n\nPushed: 2026-06-10T17:19:02Z\n\nDefault branch: develop-6.5\n\nFork: yes\n\nParent repository: openbmc/linux\n\nArchived: no\n\nREADME:\nLinux kernel\n============\n\nThere are several guides for kernel developers and users. These guides can\nbe rendered in a number of formats, like HTML and PDF. Please read\nDocumentation/admin-guide/README.rst first.\n\nIn order to build the documentation, use ``make htmldocs`` or\n``make pdfdocs``. The formatted documentation can also be read online at:\n\nhttps://www.kernel.org/doc/html/latest/\n\nThere are various text files in the Documentation/ subdirectory,\nseveral of them using the Restructured Text markup notation.\n\nPlease read the Documentation/process/changes.rst file, as it contains the\nrequirements for building and running the kernel, and information about\nthe problems which may result by upgrading your kernel."},{"ref":"P8","kind":"page","title":"NVIDIA/obmc-console repository metadata","date":"2026-06-11T07:04:12.549542+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/obmc-console","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/obmc-console\n\nDescription: OpenBMC host console infrastructure\n\nLanguage: C\n\nLicense: Apache-2.0\n\nStars: 4\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:33Z\n\nPushed: 2026-06-10T18:22:47Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/obmc-console\n\nArchived: no\n\nREADME:\n# obmc-console\n\n## To Build\n\nTo build this project, run the following shell commands:\n\nmeson setup build\nmeson compile -C build\n\nTo test:\n\ndbus-run-session meson test -C build\n\n## To Run Server\n\nRunning the server requires a serial port (e.g. /dev/ttyS0):\n\ntouch obmc-console.conf\n./obmc-console-server --config obmc-console.conf ttyS0\n\n## To Connect Client\n\nTo connect to the server, simply run the client:\n\n./obmc-console-client\n\nTo disconnect the client, use the standard `~.` combination.\n\n## Underlying design\n\nThis shows how the host UART connection is abstracted within the BMC as a Unix\ndomain socket.\n\n+---------------------------------------------------------------------------------------------+\n| |\n| obmc-console-client unix domain socket obmc-console-server |\n| |\n| +----------------------+ +------------------------+ |\n| | client.2200.conf | +---------------------+ | server.ttyVUART0.conf | |\n+---+--+ +----------------------+ | | +------------------------+ +--------+-------+\nNetwork | 2200 +--> +->+ @obmc-console.host0 +<-+ <--+ /dev/ttyVUART0 | UARTs\n+---+--+ | console-id = \"host0\" | | | | console-id = \"host0\" | +--------+-------+\n| | | +---------------------+ | | |\n| +----------------------+ +------------------------+ |\n| |\n| |\n| |\n+---------------------------------------------------------------------------------------------+\n\nThis supports multiple independent consoles. The `console-id` is a unique\nportion for the unix domain socket created by the obmc-console-server instance.\nThe server needs to know this because it needs to know what to name the pipe;\nthe client needs to know it as it needs to form the abstract socket name to\nwhich to connect.\n\n## Mux Support\n\nIn some hardware designs, multiple UARTS may be available behind a Mux. Please\nreference\n[docs/mux-support.md](https://github.com/openbmc/obmc-console/blob/master/docs/mux-support.md)\nin that case.\n\n#"},{"ref":"P9","kind":"page","title":"NVIDIA/phosphor-buttons repository metadata","date":"2026-06-11T07:04:12.237103+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/phosphor-buttons","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/phosphor-buttons\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 2\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:39Z\n\nPushed: 2026-06-10T18:22:27Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/phosphor-buttons\n\nArchived: no\n\nREADME:\n#phosphor - buttons\n\nPhosphor-buttons has a collection of IO event handler interfaces\nfor physical inputs which are part of OCP front panel.\n\nIt defines an individual dbus interface object for each physical\nbutton/switch inputs such as power button, reset button etc.\nEach of this button interfaces monitors it's associated io for event changes and calls\nthe respective event handlers.\n\n## Gpio defs config\nIn order to monitor a button/input interface the\nrespective gpio config details should be mentioned in the\ngpio defs json file - \"/etc/default/obmc/gpio/gpio_defs.json\"\n\n1. The button interface type name.\n2. An array consists of single or multiple\ngpio configs associated with the specific button interface.\n3. The name of the gpio line must be included\n4. The edge (rising or falling) must be specified. For instance,\nif a button is LOW when asserted, then edge would be falling.\n\n## example gpio def Json config\n\n{\n\"gpio_definitions\": [\n{\n\"name\": \"POWER_BUTTON\",\n\"gpio_name\": \"PWR_BTN_L-I\",\n\"direction\": \"falling\"\n},\n{\n\"name\": \"RESET_BUTTON\",\n\"gpio_name\": \"RST_BTN_L-I\",\n\"direction\": \"falling\"\n},\n{\n\"name\": \"ID_BTN\",\n\"gpio_name\": \"ID_BTN_N-I\",\n\"direction\": \"falling\"\n},\n{\n\"name\" : \"HOST_SELECTOR\",\n\"gpio_config\" : [\n{\n\"gpio_name\": \"AA4\",\n\"name\": \"one\",\n\"direction\": \"rising\"\n},\n{\n\"gpio_name\": \"AA5\",\n\"name\": \"two\",\n\"direction\": \"falling\"\n},\n{\n\"gpio_name\": \"AA6\",\n\"name\": \"three\",\n\"direction\": \"rising\"\n},\n{\n\"gpio_name\": \"AA7\",\n\"name\": \"four\",\n\"direction\": \"falling\"\n}\n]\n},\n}"},{"ref":"P10","kind":"page","title":"NVIDIA/phosphor-health-monitor repository metadata","date":"2026-06-11T07:04:12.16882+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/phosphor-health-monitor","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/phosphor-health-monitor\n\nLanguage: C++\n\nLicense: NOASSERTION\n\nStars: 2\n\nForks: 0\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:48Z\n\nPushed: 2026-06-10T18:22:22Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/phosphor-health-monitor\n\nArchived: no\n\nREADME: none published or not readable through the GitHub API."},{"ref":"P11","kind":"page","title":"NVIDIA/phosphor-logging repository metadata","date":"2026-06-11T07:04:11.979728+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/phosphor-logging","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/phosphor-logging\n\nDescription: Libraries for common event and logging creation.\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 2\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:47:57Z\n\nPushed: 2026-06-10T18:08:32Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/phosphor-logging\n\nArchived: no\n\nREADME:\n# phosphor-logging\n\nThe phosphor logging repository provides mechanisms for event and journal\nlogging.\n\n## Table Of Contents\n\n- [Building](#to-build)\n- [Structured Logging](#structured-logging)\n- [Event Logs](#event-logs)\n- [Event Log Extensions](#event-log-extensions)\n- [Remote Logging](#remote-logging-via-rsyslog)\n- [Boot Fail on Hardware Errors](#boot-fail-on-hardware-errors)\n- [Encoding BMC position in entry ID](#encoding-the-bmc-position-in-the-entry-id)\n\n## To Build\n\nTo build this package, do the following steps:\n\n1. meson builddir\n2. ninja -c builddir\n\n## Structured Logging\n\nphosphor-logging provides APIs to add program logging information to the\nsystemd-journal and it is preferred that this logging data is formatted in a\nstructured manner (using the facilities provided by the APIs).\n\nSee [Structured Logging](./docs/structured-logging.md) for more details on this\nAPI.\n\n## Event Logs\n\nOpenBMC event logs are a collection of D-Bus interfaces owned by\nphosphor-log-manager that reside at `/xyz/openbmc_project/logging/entry/X`,\nwhere X starts at 1 and is incremented for each new log.\n\nThe interfaces are:\n\n- [xyz.openbmc_project.Logging.Entry]\n- The main event log interface.\n- [xyz.openbmc_project.Association.Definitions]\n- Used for specifying inventory items as the cause of the event.\n- For more information on associations, see the [associations\ndoc][associations-doc].\n- [xyz.openbmc_project.Object.Delete]\n- Provides a Delete method to delete the event.\n- [xyz.openbmc_project.Software.Version]\n- Stores the code version that the error occurred on.\n\nOn platforms that make use of these event logs, the intent is that they are the\ncommon event log representation that other types of event logs can be created\nfrom. For example, there is code to convert these into both Redfish and IPMI\nevent logs, in addition to the event log extensions mentioned\n[b"},{"ref":"P12","kind":"page","title":"NVIDIA/phosphor-networkd repository metadata","date":"2026-06-11T07:04:11.403597+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/phosphor-networkd","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/phosphor-networkd\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 3\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:48:02Z\n\nPushed: 2026-06-10T18:33:29Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/phosphor-networkd\n\nArchived: no\n\nREADME:\n# phosphor-networkd\n\n## To Build\n\nTo build this package, do the following steps:\n\n```sh\n1. meson build\n2. ninja -C build\n```"},{"ref":"P13","kind":"page","title":"NVIDIA/phosphor-user-manager repository metadata","date":"2026-06-11T07:04:11.365072+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/phosphor-user-manager","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/phosphor-user-manager\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 2\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-05-06T16:48:15Z\n\nPushed: 2026-06-10T18:08:54Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/phosphor-user-manager\n\nArchived: no\n\nREADME:\n# phosphor-user-manager\n\n## To Build\n\nTo build this package, do the following steps:\n\n```sh\n1. meson build\n2. ninja -C build\n```\n\n### LDAP Configuration\n\n#### Configure LDAP\n\n```sh\ncurl -c cjar -b cjar -k -H \"Content-Type: application/json\" -X POST -d '{\"data\":[false,\"ldap://<ldap://<LDAP server ip/hostname>/\", \"<bindDN>\", \"<baseDN>\",\"<bindDNPassword>\",\"<searchScope>\",\"<serverType>\"]}'' https://$BMC_IP/xyz/openbmc_project/user/ldap/action/CreateConfig\n\n```\n\n#### NOTE\n\nIf the configured ldap server is secure then we need to upload the client\ncertificate and the CA certificate in following cases.\n\n- First time LDAP configuration.\n- Change the already configured Client/CA certificate\n\n#### Upload LDAP Client Certificate\n\n```sh\ncurl -c cjar -b cjar -k -H \"Content-Type: application/octet-stream\"\n-X PUT -T <FILE> https://<BMC_IP>/xyz/openbmc_project/certs/client/ldap\n```\n\n#### Upload CA Certificate\n\n```sh\ncurl -c cjar -b cjar -k -H \"Content-Type: application/octet-stream\"\n-X PUT -T <FILE> https://<BMC_IP>/xyz/openbmc_project/certs/authority/truststore\n```\n\n#### Clear LDAP Config\n\n```sh\ncurl -b cjar -k -H \"Content-Type: application/json\" -X POST -d '{\"data\":[]}' https://$BMC_IP/xyz/openbmc_project/user/ldap/config/action/delete\n```\n\n#### Get LDAP Config\n\n```sh\ncurl -b cjar -k https://$BMC_IP/xyz/openbmc_project/user/ldap/enumerate\n```"},{"ref":"P14","kind":"page","title":"NVIDIA/pldm repository metadata","date":"2026-06-11T07:04:11.222878+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/pldm","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/pldm\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 7\n\nForks: 3\n\nOpen issues: 2\n\nCreated: 2024-05-06T16:48:17Z\n\nPushed: 2026-06-10T18:33:37Z\n\nDefault branch: develop\n\nFork: yes\n\nParent repository: openbmc/pldm\n\nArchived: no\n\nREADME:\n# PLDM - Platform Level Data Model\n\n[![License](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](LICENSE)\n\n## Overview\n\nPLDM (Platform Level Data Model) is a key component of the OpenBMC project,\nproviding a standardized data model and message formats for various platform\nmanagement functionalities. It defines a method to manage, monitor, and control\nthe firmware and hardware of a system.\n\nThe OpenBMC PLDM project aims to implement the specifications defined by the\nDistributed Management Task Force (DMTF), allowing for interoperable management\ninterfaces across different hardware and firmware components.\n\n## Features\n\n- **Standardized Messaging:** Adheres to the DMTF's PLDM specifications,\nenabling consistent and interoperable communication between different\ncomponents.\n- **Modularity:** Supports multiple PLDM types, including base, FRU,Firmware\nupdate, Platform Monitoring and Control, and BIOS Control and Configuration.\n- **Extensibility:** Easily extendable to support new PLDM types and custom OEM\ncommands.\n- **Integration:** Seamlessly integrates with other OpenBMC components for\ncomprehensive system management.\n\n## Getting Started\n\n### Prerequisites\n\nTo build and run PLDM, you need the following dependencies:\n\n- `Meson`\n- `Ninja`\n\nAlternatively, source an OpenBMC ARM/x86 SDK.\n\n### Building\n\nTo build the PLDM project, follow these steps:\n\n```bash\nmeson setup build && meson compile -C build\n```\n\n### To run unit tests\n\nThe simplest way of running the tests is as described by the meson man page:\n\n```bash\nmeson test -C build\n```\n\nAlternatively, tests can be run in the OpenBMC CI docker container using\n[these](https://github.com/openbmc/docs/blob/master/testing/local-ci-build.md)\nsteps.\n\n### Generate coverage report\n\nUse Meson/Ninja coverage targets with coverage instrumentation enabled:\n\n```bash\n# 1) Configure with tests and coverage\nCC=gcc-13 CXX=g++-13 meson setup --wipe build -Dtests=enabled -Db_coverage=true\n\n# 2"},{"ref":"P15","kind":"page","title":"NVIDIA/nvrc repository metadata","date":"2026-06-11T07:04:11.087152+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/nvrc","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/nvrc\n\nDescription: The NVRC project provides a Rust binary that implements a simple init system for microVMs.\n\nLanguage: Rust\n\nLicense: Apache-2.0\n\nStars: 35\n\nForks: 16\n\nOpen issues: 20\n\nCreated: 2024-07-17T15:58:39Z\n\nPushed: 2026-06-10T18:13:47Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# NVRC - NVIDIA Runtime Container Init\n\n[![OpenSSF Scorecard](https://api.scorecard.dev/projects/github.com/NVIDIA/nvrc/badge)](https://scorecard.dev/viewer/?uri=github.com/NVIDIA/nvrc)\n\nA minimal init system (PID 1) for ephemeral NVIDIA GPU-enabled VMs running\nunder Kata Containers. NVRC sets up GPU drivers, configures hardware, spawns\nNVIDIA management daemons, and hands off to kata-agent for container\norchestration.\n\n## Design Philosophy\n\n**Fail Fast, Fail Hard**: NVRC is designed for ephemeral confidential VMs where\nany configuration failure should immediately terminate the VM. There are no\nrecovery mechanisms—if GPU initialization fails, the VM powers off. This\n\"panic-on-failure\" approach ensures:\n\n- **Security**: No undefined states in confidential computing environments\n- **Simplicity**: No complex error recovery logic to audit\n- **Clarity**: If it's running, it's configured correctly\n\n## Architecture\n\n```mermaid\nflowchart TD\nStart([NVRC starts as PID 1]) --> PanicHook[Set panic hook<br/>power off VM on panic]\nPanicHook --> MountFS[Mount filesystems<br/>/proc /dev /sys /run /tmp]\nMountFS --> LoopbackUp[Bring up loopback interface]\nLoopbackUp --> InitKernlog[Initialize kernel logging]\nInitKernlog --> PollSyslogOnce[Poll syslog once]\nPollSyslogOnce --> ParseKernel[Parse kernel parameters<br/>/proc/cmdline]\n\nParseKernel --> DetectMode[Detect mode]\nDetectMode --> ModeSelect{Mode?}\n\nModeSelect -->|gpu default| GPUMode[GPU Mode]\nModeSelect -->|cpu| CPUMode[CPU Mode]\nModeSelect -->|servicevm-nvl4| NVL4Mode[ServiceVM NVL4<br/>H100/H200/H800]\nModeSelect -->|servicevm-nvl5| NVL5Mode[ServiceVM NVL5<br/>B100/B200/B300]\n\nGPUMode --> GPUSteps[• Load nvidia.ko nvidia-uvm<br/>• Start nvidia-persistenced<br/>• nvidia-smi: lmc lgc pl srs<br/>• nv-hostengine dcgm-exporter<br/>• Generate CDI spec<br/>• Health checks]\n\nCPUMode --> CPUSteps[• Skip GPU initialization]\n\nNVL"},{"ref":"P16","kind":"page","title":"NVIDIA/TensorRT-Incubator repository metadata","date":"2026-06-11T07:04:10.907057+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/TensorRT-Incubator","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/TensorRT-Incubator\n\nDescription: Experimental projects related to TensorRT\n\nLanguage: MLIR\n\nStars: 125\n\nForks: 27\n\nOpen issues: 50\n\nCreated: 2024-07-31T17:51:02Z\n\nPushed: 2026-06-10T20:42:49Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# TensorRT Incubator\n\nThis repository includes experimental projects related to TensorRT.\nEach project is contained in its own subdirectory; see the READMEs\nthere for more details."},{"ref":"P17","kind":"page","title":"NVIDIA/TorchFort v0.2.0","date":"2026-06-11T07:04:10.308006+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/TorchFort/releases/tag/v0.2.0","signal_url":null,"signal_json_url":null,"text":"# v0.2.0\n\nRepository: NVIDIA/TorchFort\n\nTag: v0.2.0\n\nPublished: 2024-09-06T17:48:03Z\n\nPrerelease: yes\n\nRelease notes:\n## What's Changed\nThis release includes several major updates to TorchFort, including:\n* Enabling compilation of library with alternative compilers to NVHPC (e.g., GNU)\n* Enabling model/RL system training and inference on CPU\n* Enabling CPU-only builds without CUDA/NCCL\n* New reinforcement learning features, including PPO algorithms and on-policy algorithms\n* Improvements to build scripts\n\n### Breaking Changes\n#14 enables placing and running models/RL systems on CPU. To enable this, an additional `device` argument was added to the model/system creation APIs (e.g., `torchfort_model_create`). Please refer to the documentation for more details. \n\n### PRs included in this release\n* Enable support for complex gradient reduction in distributed cases. (https://github.com/NVIDIA/TorchFort/pull/2)\n* remove extraneous mpi call from cmake (https://github.com/NVIDIA/TorchFort/pull/4)\n* generalize cmake to build for different cuda archs (https://github.com/NVIDIA/TorchFort/pull/7)\n* remove hardcoded yaml-cpp path from CMakeLists.txt (https://github.com/NVIDIA/TorchFort/pull/5)\n* Build updates and improvements (https://github.com/NVIDIA/TorchFort/pull/10)\n* Update setup.cpp (https://github.com/NVIDIA/TorchFort/pull/11)\n* merging rl changes ( https://github.com/NVIDIA/TorchFort/pull/13)\n* Enable model training/inference on CPU or GPU devices. Enabling usage of alternative compilers to NVHPC, (https://github.com/NVIDIA/TorchFort/pull/14)\n* Tkurth/rl ppo (https://github.com/NVIDIA/TorchFort/pull/12)\n* Tkurth/rl tests (https://github.com/NVIDIA/TorchFort/pull/15)\n* Add train and inference functions for 5d Fortran arrays. (https://github.com/NVIDIA/TorchFort/pull/17)\n* Fix interface issues in Fortran module with gfortran. (https://github.com/NVIDIA/TorchFort/pull/18)\n* Enable builds without CUDA/GPU support (https://github.com/NVIDIA/TorchFort/pull/19)\n* Fixing up documentation. (https://github.com/NVIDIA/TorchFort/pull/20)\n* 0.2.0 release (https://github.com/NVIDIA/TorchFort/pull/21)\n\n**Full Changelog**: https://github.com/NVIDIA/TorchFort/compare/v0.1.0...v0.2.0"},{"ref":"P18","kind":"page","title":"NVIDIA/nvshmem repository metadata","date":"2026-06-11T07:04:10.181224+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/nvshmem","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/nvshmem\n\nDescription: NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmers to perform one-sided communication from within CUDA kernels and on CUDA streams.\n\nLanguage: C++\n\nLicense: Apache-2.0\n\nStars: 547\n\nForks: 84\n\nOpen issues: 42\n\nCreated: 2025-08-27T15:37:11Z\n\nPushed: 2026-06-11T03:29:56Z\n\nDefault branch: devel\n\nFork: no\n\nArchived: no\n\nREADME:\nNVSHMEM Overview\n****************\n\nNVSHMEM™ is a parallel programming interface based on OpenSHMEM that provides efficient and\nscalable communication for NVIDIA GPU clusters. NVSHMEM creates a global address space for\ndata that spans the memory of multiple GPUs and can be accessed with fine-grained \nGPU-initiated operations, CPU-initiated operations, and operations on CUDA® streams.\n\nQuick Links\n****************\n\nPlease see the following public links for information on building and working wih NVSHMEM:\n\n[Project Homepage](https://developer.nvidia.com/nvshmem)\n\n[Release Notes](https://docs.nvidia.com/nvshmem/release-notes-install-guide/release-notes/index.html)\n\n[Installation Guide](https://docs.nvidia.com/nvshmem/release-notes-install-guide/install-guide/index.html)\n\n[Best Practice Guide](https://docs.nvidia.com/nvshmem/release-notes-install-guide/best-practice-guide/index.html)\n\n[API Documentation](https://docs.nvidia.com/nvshmem/api/index.html)\n\n[Devzone Topic Page](https://forums.developer.nvidia.com/tag/nvshmem)\n\nThe maintainers of the NVSHMEM project can also be contacted by e-mail at nvshmem@nvidia.com\n\nContributions use a DCO sign-off flow rather than a CLA. See CONTRIBUTING.md.\n\nConfiguration file\n******************\n\nNVSHMEM options can be provided via a simple config file using `KEY=VALUE` syntax.\n\nConfig files are loaded in the following order (later files override earlier files):\n\n- `/etc/nvshmem.conf`\n- `~/.nvshmem.conf`\n- The file pointed to by `NVSHMEM_CONF_FILE`\n\nIf a key is present in any loaded config file, its value **overrides the corresponding environment\nvariable**.\n\nExample:\n\n```\n# Example /etc/nvshmem.conf file\nNVSHMEM_DEBUG=WARN\n# NVSHMEM_SOME_"},{"ref":"P19","kind":"page","title":"NVIDIA/cuDecomp v0.5.0","date":"2026-06-11T07:04:10.021925+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/cuDecomp/releases/tag/v0.5.0","signal_url":null,"signal_json_url":null,"text":"# v0.5.0\n\nRepository: NVIDIA/cuDecomp\n\nTag: v0.5.0\n\nPublished: 2025-04-08T22:36:15Z\n\nPrerelease: no\n\nRelease notes:\n### What's Changed\nThis release includes a number of major updates to cuDecomp. This release adds new features to make cuDecomp more flexible for users (more customizable memory orderings by pencil axis via new `transpose_mem_order` configuration option and support for input/output buffer padding in transpose and halo update APIs). This release also improves support for multi-node NVLINK (MNNVL) equipped clusters with opt-in support for fabric allocated cuDecomp workspace memory. Beyond this, this release includes expanded autotuning options and general improvements.\n\n### Breaking changes\n* https://github.com/NVIDIA/cuDecomp/pull/60 adds a new padding argument to several cuDecomp APIs: `cudecompGetPencilInfo`, `cudecompTranspose*`, and `cudecompHaloUpdate*` functions. This will require updates to existing C++ code and Fortran code (depending on usage). See https://github.com/NVIDIA/cuDecomp/pull/60 and documentation for more details.\n\n### Deprecations\n* The `Makefile`-based build has been removed.\n\n### PRs included in this release\n* Made it possible to include library header from pure C program (https://github.com/NVIDIA/cuDecomp/pull/40)\n* Adding Fortran version of Taylor Green example (https://github.com/NVIDIA/cuDecomp/pull/41)\n* Fix integer overflow issue with C++ TG example for large problems. (https://github.com/NVIDIA/cuDecomp/pull/42)\n* Benchmark updates (https://github.com/NVIDIA/cuDecomp/pull/43)\n* Use unique ID based NVSHMEM initialization method for newer NVSHMEM versions (https://github.com/NVIDIA/cuDecomp/pull/44)\n* Removing Makefile build support and related files. (https://github.com/NVIDIA/cuDecomp/pull/45)\n* Add missing preprocessor guards to fix compilation without NVSHMEM enabled. (https://github.com/NVIDIA/cuDecomp/pull/46)\n* Add small MPI_Alltoall after autotuning to work around MPI memory registration delaying cudaFree. (https://github.com/NVIDIA/cuDecomp/pull/47)\n* Address narrowing conversion errors/warnings. (https://github.com/NVIDIA/cuDecomp/pull/48)\n* Add new transpose_mem_order configuration argument to enable more fle"},{"ref":"P20","kind":"page","title":"NVIDIA/TorchFort v0.3.0","date":"2026-06-11T07:04:09.948194+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/TorchFort/releases/tag/v0.3.0","signal_url":null,"signal_json_url":null,"text":"# Multi Tensor, Multi Environment Support, Modernization of Dependencies\n\nRepository: NVIDIA/TorchFort\n\nTag: v0.3.0\n\nPublished: 2025-06-03T14:09:33Z\n\nPrerelease: no\n\nRelease notes:\nSummary Release Notes\nMajor Features and Enhancements\n1. Multi-Argument Model and Loss Support\n• Added full support for models and loss functions that require multiple input, label, and output tensors, as well as custom loss arguments. This is enabled via new `torchfort_train_multiarg` and `torchfort_inference_multiarg` APIs, with corresponding Fortran and C documentation and usage examples.\n• Introduced `torchfort_tensor_list` types and management functions (`create`, `destroy`, `add_tensor`) to facilitate passing multiple tensors to models and losses.\n• Expanded documentation and provided a comprehensive Fortran example (`examples/fortran/graph`) demonstrating online training on unstructured meshes with a MeshGraphNet-like model and a custom PyTorch loss function exported via TorchScript.\n2. TorchScript Loss Functions\n• Added support for loading custom loss functions from exported TorchScript modules via a new `torchscript` loss type. This allows users to implement arbitrary loss logic in Python and integrate it into TorchFort workflows.\n• Updated configuration and documentation to describe usage and options for TorchScript-based losses.\n3. Expanded Documentation and Examples\n• Significantly updated API and usage documentation to cover the new multi-argument interfaces, tensor list management, and custom loss workflows.\n• Added a detailed, reproducible example (`examples/fortran/graph`) including all necessary mesh data, configuration, model/loss generation scripts, and visualization tools.\nCore and API Changes\n4. Loss Function API Refactor\n• Refactored the internal loss interface: loss functions now accept an additional `extra_args` argument, supporting more flexible and extensible loss computations.\n• Implemented new `TorchscriptLoss` class for TorchScript integration, and updated the loss registry accordingly.\n5. Distributed and RL Improvements\n• Reinforcement learning (RL) off-policy and on-policy buffers now support local multi-environment updates, with new APIs and documentati"},{"ref":"P21","kind":"page","title":"NVIDIA/cuDecomp v0.5.1","date":"2026-06-11T07:04:09.904964+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/cuDecomp/releases/tag/v0.5.1","signal_url":null,"signal_json_url":null,"text":"# v0.5.1\n\nRepository: NVIDIA/cuDecomp\n\nTag: v0.5.1\n\nPublished: 2025-08-13T20:09:14Z\n\nPrerelease: no\n\nRelease notes:\n### What's Changed\nThis release includes a couple of new features in cuDecomp, minor performance improvements, and documentation fixes. This release adds the ability to capture packing kernels for the pipelined backends in CUDA graphs for better latency in some cases (enabled via new environment variable `CUDECOMP_ENABLE_CUDA_GRAPHS`, see https://github.com/NVIDIA/cuDecomp/pull/68 for more details). This release also adds a new performance reporting feature to report timing breakdowns of the communication and local processing time of transpose and halo operations launched in a user workload (enabled via new environment variable `CUDECOMP_ENABLE_PERFORMANCE_REPORT`, see https://github.com/NVIDIA/cuDecomp/pull/75 for more details). \n\n### Breaking changes\nNone.\n\n### Deprecations\nNone.\n\n### PRs included in this release\n* Improve packing kernel launch efficiency for pipelined backends using CUDA graphs. (https://github.com/NVIDIA/cuDecomp/pull/68)\n* Fix Fortran documentation of transpose_*_halo_extents and transpose_*_padding autotuning options. (https://github.com/NVIDIA/cuDecomp/pull/70)\n* Better guarding of calls to cudecompAlltoall and cudecompAlltoallPipelined in transpose implementation. (https://github.com/NVIDIA/cuDecomp/pull/71)\n* Upgrade C++ standard for builds to C++17 for libcu++ compatibility. (https://github.com/NVIDIA/cuDecomp/pull/72)\n* Improve initial NVSHMEM team synchronization for transpose ops. (https://github.com/NVIDIA/cuDecomp/pull/74)\n* Improvements to NCCL communicator management. (https://github.com/NVIDIA/cuDecomp/pull/73)\n* Fix non-nvshmem builds. (https://github.com/NVIDIA/cuDecomp/pull/76)\n* Use MPI_Alltoall instead of MPI_Alltoallv if able. (https://github.com/NVIDIA/cuDecomp/pull/77)\n* Fix undefined NVML_GPU_FABRIC_UUID_LEN usage for builds against older CUDA toolkits. (https://github.com/NVIDIA/cuDecomp/pull/78)\n* Add basic CI for compilation testing and code format checks. (https://github.com/NVIDIA/cuDecomp/pull/79)\n* Pin clang format version for CI. (https://github.com/NVIDIA/cuDecomp/pull/80)\n* Add performance repor"},{"ref":"P22","kind":"page","title":"NVIDIA/open-nvdebug repository metadata","date":"2026-06-11T07:04:09.838144+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/open-nvdebug","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/open-nvdebug\n\nDescription: Tool to collect debug logs from NVIDIA server components, in band and out-of-band.\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 6\n\nForks: 0\n\nOpen issues: 0\n\nCreated: 2025-10-02T15:44:51Z\n\nPushed: 2026-06-11T05:57:30Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# OPEN-NVDEBUG\n\n> SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.\n>\n> SPDX-License-Identifier: Apache-2.0\n\n## Description\n**open-nvdebug** is NVIDIA's comprehensive diagnostic collection tool that gathers system information from NVIDIA server platforms to troubleshoot issues effectively. It collects data through multiple methods including Out-of-Band (OOB) access via BMC and In-Band (IB) access via host systems using Redfish, SSH, and IPMI protocols.\n\n## Features\n- **Comprehensive Data Collection**: Gathers logs from multiple sources in a single command\n- **Out-of-Band (OOB)**: Remote collection via BMC using Redfish and IPMI\n- **In-Band (IB)**: Direct collection from host operating system via SSH\n- **Combined Mode**: Simultaneous OOB and IB collection for complete diagnostics\n- **Multi-Protocol Support**:\n- **Redfish API**: BMC log collection via Redfish interface\n- **SSH**: Direct SSH access to BMC and host systems\n- **IPMI**: IPMI-over-LAN for BMC communication\n- **Broad Platform Support**: Supports NVIDIA HGX™, MGX™, GB series, GH series, and Workstation platforms\n- **Automated Platform Detection**: Automatically detects baseboard type and platform architecture\n- **Remote & Local Operation**: Works from remote machines or directly on the target system\n- **Standardized Output**: Generates structured logs with HTML reports for easy analysis\n- **Parallel Collection**: Optimized multi-threaded collection for faster performance\n- **Configurable Collectors**: Spreadsheet-driven collector definitions for easy customization\n\n## Prerequisites\n\nBefore you begin, ensure you have met the following requirements:\n\n### Client Host Requirements\n- **Operating System**: Linux-based OS (Ubuntu 24.04 recommended, Ubuntu 20.04+ supported)\n- **Kernel**: Linux Kernel 4.4 or later (4.15+ recommended)\n- **Python**: Python 3.12 (re"},{"ref":"P23","kind":"page","title":"NVIDIA/dgx-spark-playbooks repository metadata","date":"2026-06-11T07:04:09.554804+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/dgx-spark-playbooks","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/dgx-spark-playbooks\n\nDescription: Collection of step-by-step playbooks for setting up AI/ML workloads on NVIDIA DGX Spark devices with Blackwell architecture.\n\nLanguage: Jupyter Notebook\n\nLicense: Apache-2.0\n\nStars: 943\n\nForks: 218\n\nOpen issues: 50\n\nCreated: 2025-10-03T18:45:19Z\n\nPushed: 2026-06-11T01:11:47Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n<p align=\"center\">\n<img src=\"src/images/dgx-spark-banner.png\" alt=\"NVIDIA DGX Spark\"/>\n</p>\n\n# DGX Spark Playbooks\n\nCollection of step-by-step playbooks for setting up AI/ML workloads on NVIDIA DGX Spark devices with Blackwell architecture.\n\n## About\n\nThese playbooks provide detailed instructions for:\n- Installing and configuring popular AI frameworks\n- Running inference with optimized models\n- Setting up development environments\n- Connecting and managing your DGX Spark device\n\nEach playbook includes prerequisites, step-by-step instructions, troubleshooting guidance, and example code.\n\n## Available Playbooks\n\n### NVIDIA\n\n- [CLI Coding Agent](nvidia/cli-coding-agent/)\n- [Comfy UI](nvidia/comfy-ui/)\n- [Connect Three DGX Spark in a Ring Topology](nvidia/connect-three-sparks/)\n- [Set Up Local Network Access](nvidia/connect-to-your-spark/)\n- [Connect Two Sparks](nvidia/connect-two-sparks/)\n- [CUDA-X Data Science](nvidia/cuda-x-data-science/)\n- [cuTile Kernels](nvidia/cutile-kernels/)\n- [DGX Dashboard](nvidia/dgx-dashboard/)\n- [FLUX.1 Dreambooth LoRA Fine-tuning](nvidia/flux-finetuning/)\n- [Run Hermes Agent with Local Models](nvidia/hermes-agent/)\n- [Install and Use Isaac Sim and Isaac Lab](nvidia/isaac/)\n- [Optimized JAX](nvidia/jax/)\n- [Live VLM WebUI](nvidia/live-vlm-webui/)\n- [Run models with llama.cpp on DGX Spark](nvidia/llama-cpp/)\n- [LLaMA Factory](nvidia/llama-factory/)\n- [LM Studio on DGX Spark](nvidia/lm-studio/)\n- [Build and Deploy a Multi-Agent Chatbot](nvidia/multi-agent-chatbot/)\n- [Multi-modal Inference](nvidia/multi-modal-inference/)\n- [Connect Multiple DGX Spark through a Switch](nvidia/multi-sparks-through-switch/)\n- [NCCL for Two Sparks](nvidia/nccl/)\n- [Fine-tune with NeMo](nvidia/nemo-fine-tune/)\n- [Run NemoClaw with a Local LLM](nvidia/nemoclaw/)\n- [🦞 Set Up Example NemoClaw Ag"},{"ref":"P24","kind":"page","title":"NVIDIA/mctp repository metadata","date":"2026-06-11T07:04:09.459198+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/mctp","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/mctp\n\nDescription: Userspace tools for MCTP stack management\n\nLanguage: C\n\nLicense: GPL-2.0\n\nStars: 5\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2025-11-04T17:41:45Z\n\nPushed: 2026-06-10T18:07:55Z\n\nDefault branch: develop\n\nFork: no\n\nArchived: no\n\nREADME:\nmctp: Userspace tools for MCTP stack management\n===============================================\n\nThis project contains two utilities for running a MCTP network from the local\nmachine:\n\n- `mctp`: A small command-line utility to query and manage the state of the\nkernel MCTP stack, in a similar way to iproute2's `ip` utility.\n\n- `mctpd`: A daemon implementing the MCTP control protocol; you'll need this\nfor the local host to perform as a bus-owner. The main function of `mctpd`\nis to assign EIDs to remote endpoints, and manage the resulting routes and\nneighbour-table entries for those endpoints.\n\nBuilding & installing\n---------------------\n\nThis project uses meson for building. To configure and compile:\n\n$ meson setup obj\n$ ninja -C obj\n\nto install to the default prefix (/usr/local), with optional `DESTDIR`:\n\n$ meson install -C obj\n\nFor integration with systemd, there are a few example configuration files and\nsystemd target definitions under the `conf/` directory. These are not installed\nby default.\n\nBy default, `meson` is configured to enable tests, which requires a few extra\ndependencies (mainly `pytest`, python libraries, and `dbus-run-session`). In\ncases where the tests are not required, you can avoid these dependencies by\nconfiguring the build tree with `-Dtests=false`:\n\n$ meson setup obj -Dtests=false\n\n`mctp` Usage\n-------------\n\nUse `mctp help` for the list of available commands:\n\n$ mctp help\nmctp link\nmctp link show [ifname]\nmctp link set <ifname> [up|down] [mtu <mtu>] [network <net>] [bus-owner <physaddr>]\n\nmctp address\nmctp address show [IFNAME]\nmctp address add <eid> dev <IFNAME>\nmctp address del <eid> dev <IFNAME>\n\nmctp route\nmctp route show [net <network>]\nmctp route add <eid>[-<eid>] via <dev> [mtu <mtu>]\nmctp route add <eid>[-<eid>] gw <eid> [net <net>] [mtu <mtu>]\nmctp route del <eid>[-<eid>] via <dev>\nmctp route del <eid>[-<eid>] gw <eid> [net <net>]\n\nmctp neigh\nmctp neigh show [dev <network>]\nmctp "},{"ref":"P25","kind":"page","title":"NVIDIA/TorchFort v0.3.1","date":"2026-06-11T07:04:09.442765+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/TorchFort/releases/tag/v0.3.1","signal_url":null,"signal_json_url":null,"text":"# v0.3.1\n\nRepository: NVIDIA/TorchFort\n\nTag: v0.3.1\n\nPublished: 2025-09-15T20:14:55Z\n\nPrerelease: no\n\nRelease notes:\n### What's Changed\nThis version of TorchFort contains a few minor updates and bug fixes. This version addresses some deficiencies in the existing Fortran interfaces, adds 64-bit integer handling to the NCCL/MPI interfaces used for distributed training and fixes a bug where the extra loss argument tensor for custom loss functions was not automatically moved to the model device. Other minor improvements are removing the dependency on HDF5 in the Fortran examples to simplify builds and adding a missing compiler define causing the cart pole example to unconditionally run on CPU.\n\n### Breaking changes\nNone.\n\n### Deprecations\nNone.\n\n### Notable PRs included in this release\n* Fortran interface fixes (https://github.com/NVIDIA/TorchFort/pull/56)\n* adding kLong support to TorchFort MPI and NCCL wrappers (https://github.com/NVIDIA/TorchFort/pull/58, https://github.com/NVIDIA/TorchFort/pull/59)\n* Add missing ENABLE_GPU define to cart pole example build. (https://github.com/NVIDIA/TorchFort/pull/62)\n* Add missing calls to move loss module and extra_loss_args to model device. (https://github.com/NVIDIA/TorchFort/pull/63)\n* Removing HDF5 dependency from examples to simplify builds. (https://github.com/NVIDIA/TorchFort/pull/66)\n\n**Full Changelog**: https://github.com/NVIDIA/TorchFort/compare/v0.3.0...v0.3.1"},{"ref":"P26","kind":"page","title":"NVIDIA/sandbox-device-plugin repository metadata","date":"2026-06-11T07:04:09.211564+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/sandbox-device-plugin","signal_url":null,"signal_json_url":null,"text":"# NVIDIA/sandbox-device-plugin\n\nDescription: Kubernetes Device Plugin to help cold plug vfio/iommufd GPUs in Kata VMs for Confidential Containers\n\nLanguage: Go\n\nLicense: BSD-3-Clause\n\nStars: 9\n\nForks: 10\n\nOpen issues: 1\n\nCreated: 2025-11-04T19:55:20Z\n\nPushed: 2026-06-10T18:01:48Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# NVIDIA K8s Device Plugin to assign Passthrough GPUs to Kata VMs for Confidential Containers\n\n## Table of Contents\n- [About](#about)\n- [Features](#features)\n- [Prerequisites](#prerequisites)\n- [Quick Start](#quick-start)\n- [Docs](#docs)\n\n## About\nThis is a kubernetes device plugin that can discover and expose GPUs for passthrough on a kubernetes node. This device plugin will enable to launch GPU attached [Kata](https://katacontainers.io/) VM based containers in your kubernetes cluster. Its specifically developed to serve Kata workloads in a Kubernetes cluster.\n\n## Features\n- Discovers Nvidia GPUs which are bound to VFIO-PCI driver and exposes them as devices available to be attached to VM in pass through mode.\n- Performs basic health check on the GPU on a kubernetes node.\n\n## Prerequisites\n- Need to have Nvidia GPU configured for GPU passthrough. Quickstart section provides details about this\n- Kubernetes version >= v1.11\n- Kata release >= v3.23.0\n\n## Quick Start\n\nBefore starting the device plug, the GPUs on a kubernetes node need to configured to be in GPU pass through mode.\n\n### Preparing a GPU to be used in pass through mode\nGPU needs to be loaded with VFIO-PCI driver to be used in pass through mode\n\n##### 1. Enable IOMMU and blacklist nouveau driver on KVM Host\n\nAppend \"**intel_iommu=on modprobe.blacklist=nouveau**\" to \"**GRUB_CMDLINE_LINUX**\" \n```shell\n$ vi /etc/default/grub\n# line 6: add (if AMD CPU, add [amd_iommu=on])\nGRUB_TIMEOUT=5\nGRUB_DISTRIBUTOR=\"$(sed 's, release .*$,,g' /etc/system-release)\"\nGRUB_DEFAULT=saved\nGRUB_DISABLE_SUBMENU=true\nGRUB_TERMINAL_OUTPUT=\"console\"\nGRUB_CMDLINE_LINUX=\"rd.lvm.lv=centos/root rd.lvm.lv=centos/swap rhgb quiet intel_iommu=on modprobe.blacklist=nouveau\"\nGRUB_DISABLE_RECOVERY=\"true\"\n```\n###### Legacy Mode (BIOS)\n```shell \ngrub2-mkconfig -o /boot/grub2/grub.cfg\nreboot\n```\n###### UEFI Mode\n``"},{"ref":"P27","kind":"page","title":"NVIDIA/nvshmem v3.4.5-0","date":"2026-06-11T07:04:08.969669+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/nvshmem/releases/tag/v3.4.5-0","signal_url":null,"signal_json_url":null,"text":"# NVSHMEM 3.4.5-0\n\nRepository: NVIDIA/nvshmem\n\nTag: v3.4.5-0\n\nPublished: 2025-10-07T18:50:29Z\n\nPrerelease: no\n\nRelease notes:\n# NVIDIA® NVSHMEM 3.4.5 Release Notes \n\nNVSHMEM is an implementation of the OpenSHMEM specification for NVIDIA GPUs. The NVSHMEM programming interface implements a Partitioned Global Address Space (PGAS) model across a cluster of NVIDIA GPUs. NVSHMEM provides an easy-to-use interface to allocate memory that is symmetrically distributed across the GPUs. In addition to a CPU-side interface, NVSHMEM provides a NVIDIA<sup>®</sup> CUDA<sup>®</sup> kernel-side interface that allows CUDA threads to access any location in the symmetrically-distributed memory. \n\nThe release notes describe the key features, software enhancements and improvements, and known issues for NVSHMEM 3.4.5 and earlier releases.\n\n## Key Features and Enhancements \n\nThis NVSHMEM release includes the following key features and enhancements:\n\n- Added support for data direct NIC configurations in the IB transports. Added a new environment variable, `NVSHMEM_DISABLE_DATA_DIRECT`, to force disable data direct NIC even when present.\n- Added support for CPU-Assisted IBGDA without the use of GDRCopy or the x86 regkey setting. \nSystems not supporting the other methods will automatically fall back to this new method.\nIt enables the use of IBGDA on a broad range of systems without the need for administrator intervention.\n- Added a new environment variable `NVSHMEM_HCA_PREFIX` to enable IB transports on systems which \nname their HCA devices in a non-standard way (for example, `ipb*` instead of `mlx5*`).\n- Deprecated support for the combined `libnvshmem.a` host and device static library.\n\n## Compatibility\n\nNVSHMEM 3.4.5 has been tested with the following: \n\nCUDA Toolkit: \n\n- 12.2\n- 12.6\n- 12.9\n- 13.0\n\nCPUs:\n\n- *x86* and NVIDIA Grace™ processors\n\nGPUs:\n\n- NVIDIA Ampere A100\n- NVIDIA Hopper™\n- NVIDIA Blackwell<sup>®</sup>\n\n## Limitations \n\n- NVSHMEM is not compatible with the PMI client library on Cray systems,\nand *must* use the NVSHMEM internal PMI-2 client library. \n- You can launch jobs with the PMI bootstrap by specifying `--mpi=pmi2`\nto Slurm and `NVSHMEM_BOOTSTRAP_PMI=PMI-2`, or direc"},{"ref":"P28","kind":"page","title":"NVIDIA/nvrc v0.0.1","date":"2026-06-11T07:04:08.872612+00:00","date_source":null,"source_url":"https://github.com/NVIDIA/nvrc/releases/tag/v0.0.1","signal_url":null,"signal_json_url":null,"text":"# Release v0.0.1\n\nRepository: NVIDIA/nvrc\n\nTag: v0.0.1\n\nPublished: 2025-10-13T18:48:27Z\n\nPrerelease: no\n\nRelease notes:\n# Verify NVRC release artifacts\n\nThis guide shows how to verify the **tarball**, the **extracted binary**, the **SBOM**, and the **SLSA provenance** for a given release.\nReleases ship **two flavors** per target:\n- `NVRC` (standard)\n- `NVRC-confidential` (built with `--features=confidential`)\n\n## Prerequisites\n- [Cosign](https://docs.sigstore.dev/) v2+ (signature & bundle verification)\n- [SLSA Verifier](https://github.com/slsa-framework/slsa-verifier) v2.7.1+\n- (Optional) [GitHub CLI](https://cli.github.com/) `gh` to download assets\n\n---\n\n## 1) Set variables\n\n```bash\n# Choose your flavor (binary name prefix) and target:\nexport BIN=NVRC # or: NVRC-confidential\nexport TARGET=x86_64-unknown-linux-musl # or: aarch64-unknown-linux-musl\n\n# Set the release tag and repository:\nexport TAG=\"vX.Y.Z\"\nexport REPO=\"owner/repo\"\n```\n\n**Example:**\n```bash\nexport BIN=NVRC-confidential\nexport TARGET=x86_64-unknown-linux-musl\nexport TAG=\"v0.0.1\"\nexport REPO=\"NVIDIA/nvrc\"\n```\n\n---\n\n## 2) Download the release assets\n\nThe release contains a **single tarball** per (flavor, target) and a per-artifact **SLSA provenance** file.\n\n```bash\ngh release download \"$TAG\" --repo \"$REPO\" --pattern \"${BIN}-${TARGET}.tar.xz\" --pattern \"${BIN}-${TARGET}.tar.xz.*\" --pattern \"${BIN}-${TARGET}.intoto.jsonl\" --dir .\n```\n\nYou should now have:\n```\n${BIN}-${TARGET}.tar.xz\n${BIN}-${TARGET}.tar.xz.sig\n${BIN}-${TARGET}.tar.xz.cert\n${BIN}-${TARGET}.tar.xz.bundle.json\n\n${BIN}-${TARGET}.intoto.jsonl\n```\n\n---\n\n## 3) Verify the **tarball**\n\nThese commands assert the tarball was signed by this repository’s **GitHub Actions** workflow on **main** and is recorded in Rekor.\n(Online verification talks to Rekor; offline verification uses the embedded bundle.)\n\n```bash\n# Online verification (Rekor)\ncosign verify-blob --rekor-url https://rekor.sigstore.dev --certificate \"${BIN}-${TARGET}.tar.xz.cert\" --signature \"${BIN}-${TARGET}.tar.xz.sig\" --certificate-identity-regexp \"^https://github.com/$REPO/.github/workflows/.+@refs/heads/main$\" --certificate-oidc-issuer \"https://token.actions.githubusercontent.com\" "},{"ref":"E1","kind":"event","title":"nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16","date":"2026-06-03T14:50:04+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16","signal_url":"https://onlylabs.fyi/signals/bc55d32f-cbc2-4c84-9214-2240adbb8b5d","signal_json_url":"https://onlylabs.fyi/signals/bc55d32f-cbc2-4c84-9214-2240adbb8b5d/signal.json","text":"model_released · nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 · signal_desk=releases · occurred_at=2026-06-03T14:50:04+00:00 · url=https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 · hf_downloads=59066 · hf_likes=189 · hf_params=560524578816 · pipeline=text-generation · license=other"},{"ref":"E2","kind":"event","title":"nvidia/Cosmos3-Super-Text2Image","date":"2026-05-28T03:49:21+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/Cosmos3-Super-Text2Image","signal_url":"https://onlylabs.fyi/signals/b849237f-e7d0-4bd7-9a38-3a69269c2e12","signal_json_url":"https://onlylabs.fyi/signals/b849237f-e7d0-4bd7-9a38-3a69269c2e12/signal.json","text":"model_released · nvidia/Cosmos3-Super-Text2Image · signal_desk=releases · occurred_at=2026-05-28T03:49:21+00:00 · url=https://huggingface.co/nvidia/Cosmos3-Super-Text2Image · hf_downloads=16542 · hf_likes=131 · hf_params=64593861104 · pipeline=text-to-image · license=other"},{"ref":"E3","kind":"event","title":"nvidia/dvlt","date":"2026-05-29T19:43:40+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/dvlt","signal_url":"https://onlylabs.fyi/signals/c573623c-e47f-4589-8242-cc85a34e63bf","signal_json_url":"https://onlylabs.fyi/signals/c573623c-e47f-4589-8242-cc85a34e63bf/signal.json","text":"model_released · nvidia/dvlt · signal_desk=releases · occurred_at=2026-05-29T19:43:40+00:00 · url=https://huggingface.co/nvidia/dvlt · hf_downloads=319 · hf_likes=37 · hf_params=117081731 · pipeline=image-to-3d · license=other · raw={\"derived_reason\":\"first-party-finetune\"}"},{"ref":"E4","kind":"event","title":"nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-Base-BF16","date":"2026-06-03T14:47:17+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-Base-BF16","signal_url":"https://onlylabs.fyi/signals/4e3dd8fe-d991-48bc-9dee-b6e55afdd1bb","signal_json_url":"https://onlylabs.fyi/signals/4e3dd8fe-d991-48bc-9dee-b6e55afdd1bb/signal.json","text":"model_released · nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-Base-BF16 · signal_desk=releases · occurred_at=2026-06-03T14:47:17+00:00 · url=https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-Base-BF16 · hf_downloads=1643 · hf_likes=25 · hf_params=560524578816 · pipeline=text-generation · license=other"},{"ref":"E5","kind":"event","title":"nvidia/4D-RGPT-8B","date":"2026-06-02T03:33:43+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/4D-RGPT-8B","signal_url":"https://onlylabs.fyi/signals/1d2445bc-e123-4451-bc89-d98304f65de6","signal_json_url":"https://onlylabs.fyi/signals/1d2445bc-e123-4451-bc89-d98304f65de6/signal.json","text":"model_released · nvidia/4D-RGPT-8B · signal_desk=releases · occurred_at=2026-06-02T03:33:43+00:00 · url=https://huggingface.co/nvidia/4D-RGPT-8B · hf_downloads=154 · hf_likes=15 · pipeline=video-text-to-text · license=cc-by-nc-4.0 · raw={\"derived_reason\":\"first-party-finetune\"}"},{"ref":"E6","kind":"event","title":"nvidia/ArtiFixer","date":"2026-06-04T06:03:25+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/ArtiFixer","signal_url":"https://onlylabs.fyi/signals/49ddc4cb-cbd3-4d8e-8877-598b24ec4fa9","signal_json_url":"https://onlylabs.fyi/signals/49ddc4cb-cbd3-4d8e-8877-598b24ec4fa9/signal.json","text":"model_released · nvidia/ArtiFixer · signal_desk=releases · occurred_at=2026-06-04T06:03:25+00:00 · url=https://huggingface.co/nvidia/ArtiFixer · hf_likes=12"},{"ref":"E7","kind":"event","title":"nvidia/NV-OneFormer","date":"2026-05-29T02:14:04+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/NV-OneFormer","signal_url":"https://onlylabs.fyi/signals/a7644c32-b3f4-40f0-beda-8e8d8ec76aff","signal_json_url":"https://onlylabs.fyi/signals/a7644c32-b3f4-40f0-beda-8e8d8ec76aff/signal.json","text":"model_released · nvidia/NV-OneFormer · signal_desk=releases · occurred_at=2026-05-29T02:14:04+00:00 · url=https://huggingface.co/nvidia/NV-OneFormer · hf_likes=9"},{"ref":"E8","kind":"event","title":"NVIDIA/open-nvdebug nvdebug-v2.1.0-release","date":"2026-06-11T05:57:30+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/open-nvdebug/releases/tag/nvdebug-v2.1.0-release","signal_url":"https://onlylabs.fyi/signals/45f4cdf4-5665-44fd-b290-c27312411e95","signal_json_url":"https://onlylabs.fyi/signals/45f4cdf4-5665-44fd-b290-c27312411e95/signal.json","text":"release · NVIDIA/open-nvdebug nvdebug-v2.1.0-release · signal_desk=releases · occurred_at=2026-06-11T05:57:30+00:00 · url=https://github.com/NVIDIA/open-nvdebug/releases/tag/nvdebug-v2.1.0-release · raw={\"repo\":\"NVIDIA/open-nvdebug\"}"},{"ref":"E9","kind":"event","title":"NVIDIA/open-nvdebug nvdebug-v2.0.0","date":"2026-06-11T05:55:39+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/open-nvdebug/releases/tag/nvdebug-v2.0.0","signal_url":"https://onlylabs.fyi/signals/25a9df96-2996-450c-a19b-ac955b7e77b7","signal_json_url":"https://onlylabs.fyi/signals/25a9df96-2996-450c-a19b-ac955b7e77b7/signal.json","text":"release · NVIDIA/open-nvdebug nvdebug-v2.0.0 · signal_desk=releases · occurred_at=2026-06-11T05:55:39+00:00 · url=https://github.com/NVIDIA/open-nvdebug/releases/tag/nvdebug-v2.0.0 · raw={\"repo\":\"NVIDIA/open-nvdebug\"}"},{"ref":"E10","kind":"event","title":"NVIDIA/nv-rms-client v0.9.0-mts-rc02","date":"2026-06-11T05:19:43+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/nv-rms-client/releases/tag/v0.9.0-mts-rc02","signal_url":"https://onlylabs.fyi/signals/837ce80c-4bdb-48fc-947b-8842ace3a199","signal_json_url":"https://onlylabs.fyi/signals/837ce80c-4bdb-48fc-947b-8842ace3a199/signal.json","text":"release · NVIDIA/nv-rms-client v0.9.0-mts-rc02 · signal_desk=releases · occurred_at=2026-06-11T05:19:43+00:00 · url=https://github.com/NVIDIA/nv-rms-client/releases/tag/v0.9.0-mts-rc02 · raw={\"repo\":\"NVIDIA/nv-rms-client\"}"},{"ref":"E11","kind":"event","title":"NVIDIA/cudnn-frontend v1.25.0","date":"2026-06-10T21:11:51+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/cudnn-frontend/releases/tag/v1.25.0","signal_url":"https://onlylabs.fyi/signals/1f40b37a-1570-4da9-b4ed-a722da3acd28","signal_json_url":"https://onlylabs.fyi/signals/1f40b37a-1570-4da9-b4ed-a722da3acd28/signal.json","text":"release · NVIDIA/cudnn-frontend v1.25.0 · signal_desk=releases · occurred_at=2026-06-10T21:11:51+00:00 · url=https://github.com/NVIDIA/cudnn-frontend/releases/tag/v1.25.0 · raw={\"repo\":\"NVIDIA/cudnn-frontend\"}"},{"ref":"E12","kind":"event","title":"NVIDIA/nv-rms-client v0.9.0-mts-rc01","date":"2026-06-10T19:02:10+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/nv-rms-client/releases/tag/v0.9.0-mts-rc01","signal_url":"https://onlylabs.fyi/signals/c5fe1a74-96b4-4f19-a620-d1b63758cdcb","signal_json_url":"https://onlylabs.fyi/signals/c5fe1a74-96b4-4f19-a620-d1b63758cdcb/signal.json","text":"release · NVIDIA/nv-rms-client v0.9.0-mts-rc01 · signal_desk=releases · occurred_at=2026-06-10T19:02:10+00:00 · url=https://github.com/NVIDIA/nv-rms-client/releases/tag/v0.9.0-mts-rc01 · raw={\"repo\":\"NVIDIA/nv-rms-client\"}"},{"ref":"E13","kind":"event","title":"For Robotaxis, Safety Must Be Built In, Not Bolted On","date":"2026-06-10T19:00:12+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/halos-os-robotaxi-safety/","signal_url":"https://onlylabs.fyi/signals/9f47eb49-25ba-4e10-9edb-b1e747b43248","signal_json_url":"https://onlylabs.fyi/signals/9f47eb49-25ba-4e10-9edb-b1e747b43248/signal.json","text":"post_published · For Robotaxis, Safety Must Be Built In, Not Bolted On · signal_desk=talking · occurred_at=2026-06-10T19:00:12+00:00 · url=https://blogs.nvidia.com/blog/halos-os-robotaxi-safety/ · data_radar_lanes=Safety and policy · data_radar_terms=safety · data_radar_reason=NVIDIA has a writing signal matching safety and policy. · raw={\"excerpt\":\"A car pulls up to the curb. The app says, “Your ride is here.” No one’s in the driver’s seat. For people who live in one of the dozens of cities now hosting robotaxi services, this is already a reality. The robotaxi industry has moved from prototype milestones to commercial operations, with an expanding ecosystem […]\"}"},{"ref":"E14","kind":"event","title":"NVIDIA/elements @nvidia-elements/styles-v2.0.2","date":"2026-06-10T18:51:17+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/styles-v2.0.2","signal_url":"https://onlylabs.fyi/signals/565de597-ad94-49ff-91ef-632e549ba88d","signal_json_url":"https://onlylabs.fyi/signals/565de597-ad94-49ff-91ef-632e549ba88d/signal.json","text":"release · NVIDIA/elements @nvidia-elements/styles-v2.0.2 · signal_desk=releases · occurred_at=2026-06-10T18:51:17+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/styles-v2.0.2 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E15","kind":"event","title":"NVIDIA/attestation-sdk 2026.06.09","date":"2026-06-10T17:23:54+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/attestation-sdk/releases/tag/2026.06.09","signal_url":"https://onlylabs.fyi/signals/fc7bd045-f90d-4df7-898d-0b2e2fdce1df","signal_json_url":"https://onlylabs.fyi/signals/fc7bd045-f90d-4df7-898d-0b2e2fdce1df/signal.json","text":"release · NVIDIA/attestation-sdk 2026.06.09 · signal_desk=releases · occurred_at=2026-06-10T17:23:54+00:00 · url=https://github.com/NVIDIA/attestation-sdk/releases/tag/2026.06.09 · raw={\"repo\":\"NVIDIA/attestation-sdk\"}"},{"ref":"E16","kind":"event","title":"NVIDIA Accelerates Google DeepMind’s DiffusionGemma for Local AI","date":"2026-06-10T16:15:20+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/rtx-ai-garage-local-gemma-diffusion/","signal_url":"https://onlylabs.fyi/signals/cd926e0a-3925-4081-b7d5-635fb25b81af","signal_json_url":"https://onlylabs.fyi/signals/cd926e0a-3925-4081-b7d5-635fb25b81af/signal.json","text":"post_published · NVIDIA Accelerates Google DeepMind’s DiffusionGemma for Local AI · signal_desk=talking · occurred_at=2026-06-10T16:15:20+00:00 · url=https://blogs.nvidia.com/blog/rtx-ai-garage-local-gemma-diffusion/ · data_radar_lanes=Infrastructure · data_radar_terms=platform, systems, gpu · data_radar_reason=NVIDIA has a writing signal matching infrastructure. · raw={\"excerpt\":\"Today, Google DeepMind released DiffusionGemma — an experimental open model built for exceptionally fast text generation. NVIDIA has optimized DiffusionGemma to run even faster across NVIDIA GeForce RTX GPUs, the NVIDIA RTX PRO platform and NVIDIA DGX Spark systems, from local PCs to the cloud.  Rather than generating text one word at a time, DiffusionGemma generates multiple words in parallel to output whole blocks of text, opening a new, low-latency frontier for the kind of single-user workloads that developers, […]\"}"},{"ref":"E17","kind":"event","title":"NVIDIA/elements @nvidia-elements/markdown-v2.0.1","date":"2026-06-10T15:43:49+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/markdown-v2.0.1","signal_url":"https://onlylabs.fyi/signals/53ff7777-1e0e-4673-b0e2-93d9eee82f1e","signal_json_url":"https://onlylabs.fyi/signals/53ff7777-1e0e-4673-b0e2-93d9eee82f1e/signal.json","text":"release · NVIDIA/elements @nvidia-elements/markdown-v2.0.1 · signal_desk=releases · occurred_at=2026-06-10T15:43:49+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/markdown-v2.0.1 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E18","kind":"event","title":"NVIDIA/elements @nvidia-elements/code-v2.0.1","date":"2026-06-10T15:43:21+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/code-v2.0.1","signal_url":"https://onlylabs.fyi/signals/ab10d830-c17c-4144-8f71-b175c14b5198","signal_json_url":"https://onlylabs.fyi/signals/ab10d830-c17c-4144-8f71-b175c14b5198/signal.json","text":"release · NVIDIA/elements @nvidia-elements/code-v2.0.1 · signal_desk=releases · occurred_at=2026-06-10T15:43:21+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/code-v2.0.1 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E19","kind":"event","title":"NVIDIA/elements @nvidia-elements/monaco-v2.0.1","date":"2026-06-10T15:24:03+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/monaco-v2.0.1","signal_url":"https://onlylabs.fyi/signals/07e3dd6c-81fc-492a-bf16-c43f2abc45fb","signal_json_url":"https://onlylabs.fyi/signals/07e3dd6c-81fc-492a-bf16-c43f2abc45fb/signal.json","text":"release · NVIDIA/elements @nvidia-elements/monaco-v2.0.1 · signal_desk=releases · occurred_at=2026-06-10T15:24:03+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/monaco-v2.0.1 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E20","kind":"event","title":"NVIDIA/elements @nvidia-elements/styles-v2.0.1","date":"2026-06-10T15:23:32+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/styles-v2.0.1","signal_url":"https://onlylabs.fyi/signals/5e445de0-7f55-4f76-a5dc-36b46bcd7350","signal_json_url":"https://onlylabs.fyi/signals/5e445de0-7f55-4f76-a5dc-36b46bcd7350/signal.json","text":"release · NVIDIA/elements @nvidia-elements/styles-v2.0.1 · signal_desk=releases · occurred_at=2026-06-10T15:23:32+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/styles-v2.0.1 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E21","kind":"event","title":"NVIDIA/holodeck v0.3.6","date":"2026-06-10T05:41:09+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/holodeck/releases/tag/v0.3.6","signal_url":"https://onlylabs.fyi/signals/954970e0-2b11-4c4e-afef-82ebd4e52192","signal_json_url":"https://onlylabs.fyi/signals/954970e0-2b11-4c4e-afef-82ebd4e52192/signal.json","text":"release · NVIDIA/holodeck v0.3.6 · signal_desk=releases · occurred_at=2026-06-10T05:41:09+00:00 · url=https://github.com/NVIDIA/holodeck/releases/tag/v0.3.6 · raw={\"repo\":\"NVIDIA/holodeck\"}"},{"ref":"E22","kind":"event","title":"NVIDIA/TensorRT-LLM v1.3.0rc18","date":"2026-06-10T00:10:37+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/TensorRT-LLM/releases/tag/v1.3.0rc18","signal_url":"https://onlylabs.fyi/signals/ce0f4399-15b8-4217-a373-5e5ef4480d51","signal_json_url":"https://onlylabs.fyi/signals/ce0f4399-15b8-4217-a373-5e5ef4480d51/signal.json","text":"release · NVIDIA/TensorRT-LLM v1.3.0rc18 · signal_desk=releases · occurred_at=2026-06-10T00:10:37+00:00 · url=https://github.com/NVIDIA/TensorRT-LLM/releases/tag/v1.3.0rc18 · raw={\"repo\":\"NVIDIA/TensorRT-LLM\"}"},{"ref":"E23","kind":"event","title":"NVIDIA Confidential Computing to Help Expand Apple’s Private Cloud Compute","date":"2026-06-09T22:34:27+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/nvidia-confidential-computing-apple-private-cloud-compute/","signal_url":"https://onlylabs.fyi/signals/a1f83015-fdf5-4d2a-9abd-08e6d42439d8","signal_json_url":"https://onlylabs.fyi/signals/a1f83015-fdf5-4d2a-9abd-08e6d42439d8/signal.json","text":"post_published · NVIDIA Confidential Computing to Help Expand Apple’s Private Cloud Compute · signal_desk=talking · occurred_at=2026-06-09T22:34:27+00:00 · url=https://blogs.nvidia.com/blog/nvidia-confidential-computing-apple-private-cloud-compute/ · data_radar_lanes=Data demand, Infrastructure, Product and customer · data_radar_terms=data, rag, inference, gpu, support · data_radar_reason=NVIDIA has a writing signal matching data demand, infrastructure, product and customer. · raw={\"excerpt\":\"NVIDIA GPUs with Confidential Computing are now used for confidential inference in Apple’s Private Cloud Compute (PCC), as it expands beyond Apple’s data centers to Google Cloud.  Unveiled during Apple’s annual WWDC gathering for developers from around the globe, NVIDIA GPUs will support server-side inference for Apple Foundation Models, custom-built by Apple and Google, leveraging […]\"}"},{"ref":"E24","kind":"event","title":"NVIDIA/cuopt v26.06.00","date":"2026-06-09T22:05:51+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/cuopt/releases/tag/v26.06.00","signal_url":"https://onlylabs.fyi/signals/e3ba243c-f041-46d6-bcfe-f6a2d00fe5ba","signal_json_url":"https://onlylabs.fyi/signals/e3ba243c-f041-46d6-bcfe-f6a2d00fe5ba/signal.json","text":"release · NVIDIA/cuopt v26.06.00 · signal_desk=releases · occurred_at=2026-06-09T22:05:51+00:00 · url=https://github.com/NVIDIA/cuopt/releases/tag/v26.06.00 · raw={\"repo\":\"NVIDIA/cuopt\"}"},{"ref":"E25","kind":"event","title":"NVIDIA/dsx-exchange v2.5.24","date":"2026-06-09T21:40:20+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.24","signal_url":"https://onlylabs.fyi/signals/4a83bdf4-48cf-49a5-ba7d-cd2fb0d4248e","signal_json_url":"https://onlylabs.fyi/signals/4a83bdf4-48cf-49a5-ba7d-cd2fb0d4248e/signal.json","text":"release · NVIDIA/dsx-exchange v2.5.24 · signal_desk=releases · occurred_at=2026-06-09T21:40:20+00:00 · url=https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.24 · raw={\"repo\":\"NVIDIA/dsx-exchange\"}"},{"ref":"E26","kind":"event","title":"NVIDIA/dsx-exchange v2.5.23","date":"2026-06-09T21:27:48+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.23","signal_url":"https://onlylabs.fyi/signals/7ebc5af1-4c41-430e-84fd-c1cf16572f95","signal_json_url":"https://onlylabs.fyi/signals/7ebc5af1-4c41-430e-84fd-c1cf16572f95/signal.json","text":"release · NVIDIA/dsx-exchange v2.5.23 · signal_desk=releases · occurred_at=2026-06-09T21:27:48+00:00 · url=https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.23 · raw={\"repo\":\"NVIDIA/dsx-exchange\"}"},{"ref":"E27","kind":"event","title":"NVIDIA/libredfish v0.44.10","date":"2026-06-09T20:58:16+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/libredfish/releases/tag/v0.44.10","signal_url":"https://onlylabs.fyi/signals/2b150d51-c9d5-49b3-a3fd-a6b0066837fb","signal_json_url":"https://onlylabs.fyi/signals/2b150d51-c9d5-49b3-a3fd-a6b0066837fb/signal.json","text":"release · NVIDIA/libredfish v0.44.10 · signal_desk=releases · occurred_at=2026-06-09T20:58:16+00:00 · url=https://github.com/NVIDIA/libredfish/releases/tag/v0.44.10 · raw={\"repo\":\"NVIDIA/libredfish\"}"},{"ref":"E28","kind":"event","title":"NVIDIA/holodeck v0.3.5","date":"2026-06-09T20:53:02+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/holodeck/releases/tag/v0.3.5","signal_url":"https://onlylabs.fyi/signals/268d020f-1526-4658-bed6-c6d4563a9a0b","signal_json_url":"https://onlylabs.fyi/signals/268d020f-1526-4658-bed6-c6d4563a9a0b/signal.json","text":"release · NVIDIA/holodeck v0.3.5 · signal_desk=releases · occurred_at=2026-06-09T20:53:02+00:00 · url=https://github.com/NVIDIA/holodeck/releases/tag/v0.3.5 · raw={\"repo\":\"NVIDIA/holodeck\"}"},{"ref":"E29","kind":"event","title":"NVIDIA/dsx-exchange v2.5.22","date":"2026-06-09T20:38:46+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.22","signal_url":"https://onlylabs.fyi/signals/bc8bff0a-5104-466d-bdef-779ee11320af","signal_json_url":"https://onlylabs.fyi/signals/bc8bff0a-5104-466d-bdef-779ee11320af/signal.json","text":"release · NVIDIA/dsx-exchange v2.5.22 · signal_desk=releases · occurred_at=2026-06-09T20:38:46+00:00 · url=https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.22 · raw={\"repo\":\"NVIDIA/dsx-exchange\"}"},{"ref":"E30","kind":"event","title":"NVIDIA/OpenShell v0.0.59","date":"2026-06-09T20:07:17+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/OpenShell/releases/tag/v0.0.59","signal_url":"https://onlylabs.fyi/signals/d31eb249-81a4-4ddb-9b51-057132a3ae30","signal_json_url":"https://onlylabs.fyi/signals/d31eb249-81a4-4ddb-9b51-057132a3ae30/signal.json","text":"release · NVIDIA/OpenShell v0.0.59 · signal_desk=releases · occurred_at=2026-06-09T20:07:17+00:00 · url=https://github.com/NVIDIA/OpenShell/releases/tag/v0.0.59 · raw={\"repo\":\"NVIDIA/OpenShell\"}"},{"ref":"E31","kind":"event","title":"NVIDIA/dsx-exchange v2.5.21","date":"2026-06-09T18:21:09+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.21","signal_url":"https://onlylabs.fyi/signals/f70e576f-e661-4a8a-92e3-0b93b4ff1fbe","signal_json_url":"https://onlylabs.fyi/signals/f70e576f-e661-4a8a-92e3-0b93b4ff1fbe/signal.json","text":"release · NVIDIA/dsx-exchange v2.5.21 · signal_desk=releases · occurred_at=2026-06-09T18:21:09+00:00 · url=https://github.com/NVIDIA/dsx-exchange/releases/tag/v2.5.21 · raw={\"repo\":\"NVIDIA/dsx-exchange\"}"},{"ref":"E32","kind":"event","title":"NVIDIA/TensorRT-RTX-EP-ABI v0.3.0","date":"2026-06-09T16:36:14+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/TensorRT-RTX-EP-ABI/releases/tag/v0.3.0","signal_url":"https://onlylabs.fyi/signals/1018025d-f746-4ade-a798-392cbdf3b7df","signal_json_url":"https://onlylabs.fyi/signals/1018025d-f746-4ade-a798-392cbdf3b7df/signal.json","text":"release · NVIDIA/TensorRT-RTX-EP-ABI v0.3.0 · signal_desk=releases · occurred_at=2026-06-09T16:36:14+00:00 · url=https://github.com/NVIDIA/TensorRT-RTX-EP-ABI/releases/tag/v0.3.0 · raw={\"repo\":\"NVIDIA/TensorRT-RTX-EP-ABI\"}"},{"ref":"E33","kind":"event","title":"NVIDIA/cccl python-1.0.1","date":"2026-06-09T15:09:36+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/cccl/releases/tag/python-1.0.1","signal_url":"https://onlylabs.fyi/signals/9defe32b-80fc-4d37-a4a8-b4e0366070c1","signal_json_url":"https://onlylabs.fyi/signals/9defe32b-80fc-4d37-a4a8-b4e0366070c1/signal.json","text":"release · NVIDIA/cccl python-1.0.1 · signal_desk=releases · occurred_at=2026-06-09T15:09:36+00:00 · url=https://github.com/NVIDIA/cccl/releases/tag/python-1.0.1 · raw={\"repo\":\"NVIDIA/cccl\"}"},{"ref":"E34","kind":"event","title":"nvidia/GR00T-H-N1.7","date":"2026-05-30T14:40:32+00:00","date_source":"source","source_url":"https://huggingface.co/nvidia/GR00T-H-N1.7","signal_url":"https://onlylabs.fyi/signals/15c241a3-cfcf-48a6-8987-9fdfae82c900","signal_json_url":"https://onlylabs.fyi/signals/15c241a3-cfcf-48a6-8987-9fdfae82c900/signal.json","text":"model_released · nvidia/GR00T-H-N1.7 · signal_desk=releases · occurred_at=2026-05-30T14:40:32+00:00 · url=https://huggingface.co/nvidia/GR00T-H-N1.7 · hf_downloads=24 · hf_likes=5 · hf_params=2942584448 · license=other · raw={\"derived_reason\":\"first-party-finetune\"}"},{"ref":"E35","kind":"event","title":"NVIDIA/spark-rapids-tools v26.04.5","date":"2026-06-09T08:49:42+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/spark-rapids-tools/releases/tag/v26.04.5","signal_url":"https://onlylabs.fyi/signals/46b83b04-bf92-43d7-a19e-3fb0228ee30f","signal_json_url":"https://onlylabs.fyi/signals/46b83b04-bf92-43d7-a19e-3fb0228ee30f/signal.json","text":"release · NVIDIA/spark-rapids-tools v26.04.5 · signal_desk=releases · occurred_at=2026-06-09T08:49:42+00:00 · url=https://github.com/NVIDIA/spark-rapids-tools/releases/tag/v26.04.5 · raw={\"repo\":\"NVIDIA/spark-rapids-tools\"}"},{"ref":"E36","kind":"event","title":"NVIDIA/libredfish v0.44.9","date":"2026-06-09T08:17:46+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/libredfish/releases/tag/v0.44.9","signal_url":"https://onlylabs.fyi/signals/2775426f-798a-4f9b-b9cc-638e9d8ba0d7","signal_json_url":"https://onlylabs.fyi/signals/2775426f-798a-4f9b-b9cc-638e9d8ba0d7/signal.json","text":"release · NVIDIA/libredfish v0.44.9 · signal_desk=releases · occurred_at=2026-06-09T08:17:46+00:00 · url=https://github.com/NVIDIA/libredfish/releases/tag/v0.44.9 · raw={\"repo\":\"NVIDIA/libredfish\"}"},{"ref":"E37","kind":"event","title":"NVIDIA/TransformerEngine v2.16","date":"2026-06-09T01:15:56+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/TransformerEngine/releases/tag/v2.16","signal_url":"https://onlylabs.fyi/signals/243a16ed-2d56-44aa-8c92-e1e42deef8b0","signal_json_url":"https://onlylabs.fyi/signals/243a16ed-2d56-44aa-8c92-e1e42deef8b0/signal.json","text":"release · NVIDIA/TransformerEngine v2.16 · signal_desk=releases · occurred_at=2026-06-09T01:15:56+00:00 · url=https://github.com/NVIDIA/TransformerEngine/releases/tag/v2.16 · raw={\"repo\":\"NVIDIA/TransformerEngine\"}"},{"ref":"E38","kind":"event","title":"NVIDIA/enroot v4.2.1","date":"2026-06-09T00:34:35+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/enroot/releases/tag/v4.2.1","signal_url":"https://onlylabs.fyi/signals/7cf0e789-738d-4217-b869-606ad918bbc2","signal_json_url":"https://onlylabs.fyi/signals/7cf0e789-738d-4217-b869-606ad918bbc2/signal.json","text":"release · NVIDIA/enroot v4.2.1 · signal_desk=releases · occurred_at=2026-06-09T00:34:35+00:00 · url=https://github.com/NVIDIA/enroot/releases/tag/v4.2.1 · raw={\"repo\":\"NVIDIA/enroot\"}"},{"ref":"E39","kind":"event","title":"NVIDIA/physicsnemo v2.1.1","date":"2026-06-08T23:37:58+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/physicsnemo/releases/tag/v2.1.1","signal_url":"https://onlylabs.fyi/signals/42d496d9-2b20-4702-a317-efaf3b976c99","signal_json_url":"https://onlylabs.fyi/signals/42d496d9-2b20-4702-a317-efaf3b976c99/signal.json","text":"release · NVIDIA/physicsnemo v2.1.1 · signal_desk=releases · occurred_at=2026-06-08T23:37:58+00:00 · url=https://github.com/NVIDIA/physicsnemo/releases/tag/v2.1.1 · raw={\"repo\":\"NVIDIA/physicsnemo\"}"},{"ref":"E40","kind":"event","title":"NVIDIA/nccl nccl-ep-v0.1.0","date":"2026-06-08T22:54:11+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/nccl/releases/tag/nccl-ep-v0.1.0","signal_url":"https://onlylabs.fyi/signals/40d9f453-e84f-43a2-8f34-ee27a563f9db","signal_json_url":"https://onlylabs.fyi/signals/40d9f453-e84f-43a2-8f34-ee27a563f9db/signal.json","text":"release · NVIDIA/nccl nccl-ep-v0.1.0 · signal_desk=releases · occurred_at=2026-06-08T22:54:11+00:00 · url=https://github.com/NVIDIA/nccl/releases/tag/nccl-ep-v0.1.0 · raw={\"repo\":\"NVIDIA/nccl\"}"},{"ref":"E41","kind":"event","title":"NVIDIA/ais-k8s v3.0.0","date":"2026-06-08T21:45:00+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/ais-k8s/releases/tag/v3.0.0","signal_url":"https://onlylabs.fyi/signals/2091fc0e-f41f-4ab2-9990-d6693dfde2ac","signal_json_url":"https://onlylabs.fyi/signals/2091fc0e-f41f-4ab2-9990-d6693dfde2ac/signal.json","text":"release · NVIDIA/ais-k8s v3.0.0 · signal_desk=releases · occurred_at=2026-06-08T21:45:00+00:00 · url=https://github.com/NVIDIA/ais-k8s/releases/tag/v3.0.0 · raw={\"repo\":\"NVIDIA/ais-k8s\"}"},{"ref":"E42","kind":"event","title":"NVIDIA/nodewright cli/v0.2.0-rc.1","date":"2026-06-08T21:37:36+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/nodewright/releases/tag/cli/v0.2.0-rc.1","signal_url":"https://onlylabs.fyi/signals/c6b3364c-07c8-45c9-99fb-54177b0a86d9","signal_json_url":"https://onlylabs.fyi/signals/c6b3364c-07c8-45c9-99fb-54177b0a86d9/signal.json","text":"release · NVIDIA/nodewright cli/v0.2.0-rc.1 · signal_desk=releases · occurred_at=2026-06-08T21:37:36+00:00 · url=https://github.com/NVIDIA/nodewright/releases/tag/cli/v0.2.0-rc.1 · raw={\"repo\":\"NVIDIA/nodewright\"}"},{"ref":"E43","kind":"event","title":"NVIDIA/nv-redfish v0.10.2","date":"2026-06-08T20:14:22+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/nv-redfish/releases/tag/v0.10.2","signal_url":"https://onlylabs.fyi/signals/ff96dbb7-44c8-4b6d-ae19-0b2589857705","signal_json_url":"https://onlylabs.fyi/signals/ff96dbb7-44c8-4b6d-ae19-0b2589857705/signal.json","text":"release · NVIDIA/nv-redfish v0.10.2 · signal_desk=releases · occurred_at=2026-06-08T20:14:22+00:00 · url=https://github.com/NVIDIA/nv-redfish/releases/tag/v0.10.2 · raw={\"repo\":\"NVIDIA/nv-redfish\"}"},{"ref":"E44","kind":"event","title":"NVIDIA/cudnn-frontend v1.24.1","date":"2026-06-08T19:04:16+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/cudnn-frontend/releases/tag/v1.24.1","signal_url":"https://onlylabs.fyi/signals/d45e98df-635c-46f4-9a83-91c4578d6171","signal_json_url":"https://onlylabs.fyi/signals/d45e98df-635c-46f4-9a83-91c4578d6171/signal.json","text":"release · NVIDIA/cudnn-frontend v1.24.1 · signal_desk=releases · occurred_at=2026-06-08T19:04:16+00:00 · url=https://github.com/NVIDIA/cudnn-frontend/releases/tag/v1.24.1 · raw={\"repo\":\"NVIDIA/cudnn-frontend\"}"},{"ref":"E45","kind":"event","title":"NVIDIA/elements @nvidia-elements/markdown-v2.0.0","date":"2026-06-08T18:31:45+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/markdown-v2.0.0","signal_url":"https://onlylabs.fyi/signals/56babb6e-ecba-492c-9d93-74b355fb6d83","signal_json_url":"https://onlylabs.fyi/signals/56babb6e-ecba-492c-9d93-74b355fb6d83/signal.json","text":"release · NVIDIA/elements @nvidia-elements/markdown-v2.0.0 · signal_desk=releases · occurred_at=2026-06-08T18:31:45+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/markdown-v2.0.0 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E46","kind":"event","title":"NVIDIA/elements @nvidia-elements/code-v2.0.0","date":"2026-06-08T18:31:17+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/code-v2.0.0","signal_url":"https://onlylabs.fyi/signals/928d7c02-7f13-4c69-946e-08ae3ca10c90","signal_json_url":"https://onlylabs.fyi/signals/928d7c02-7f13-4c69-946e-08ae3ca10c90/signal.json","text":"release · NVIDIA/elements @nvidia-elements/code-v2.0.0 · signal_desk=releases · occurred_at=2026-06-08T18:31:17+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/code-v2.0.0 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E47","kind":"event","title":"NVIDIA/elements @nvidia-elements/cli-v2.0.0","date":"2026-06-08T18:30:52+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/cli-v2.0.0","signal_url":"https://onlylabs.fyi/signals/e78ba103-0ceb-4a0c-8074-b4310f37b35b","signal_json_url":"https://onlylabs.fyi/signals/e78ba103-0ceb-4a0c-8074-b4310f37b35b/signal.json","text":"release · NVIDIA/elements @nvidia-elements/cli-v2.0.0 · signal_desk=releases · occurred_at=2026-06-08T18:30:52+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/cli-v2.0.0 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E48","kind":"event","title":"NVIDIA/elements @nvidia-elements/monaco-v2.0.0","date":"2026-06-08T18:30:25+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/monaco-v2.0.0","signal_url":"https://onlylabs.fyi/signals/ba01459a-c740-4819-a08b-a2e8eb79bfd4","signal_json_url":"https://onlylabs.fyi/signals/ba01459a-c740-4819-a08b-a2e8eb79bfd4/signal.json","text":"release · NVIDIA/elements @nvidia-elements/monaco-v2.0.0 · signal_desk=releases · occurred_at=2026-06-08T18:30:25+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/monaco-v2.0.0 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E49","kind":"event","title":"NVIDIA/elements @nvidia-elements/lint-v2.0.0","date":"2026-06-08T18:29:57+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/lint-v2.0.0","signal_url":"https://onlylabs.fyi/signals/df37aa42-c958-4cae-bfe0-72b720bb74d6","signal_json_url":"https://onlylabs.fyi/signals/df37aa42-c958-4cae-bfe0-72b720bb74d6/signal.json","text":"release · NVIDIA/elements @nvidia-elements/lint-v2.0.0 · signal_desk=releases · occurred_at=2026-06-08T18:29:57+00:00 · url=https://github.com/NVIDIA/elements/releases/tag/%40nvidia-elements/lint-v2.0.0 · raw={\"repo\":\"NVIDIA/elements\"}"},{"ref":"E50","kind":"event","title":"NVIDIA/go-nvml v0.13.1-0","date":"2026-06-08T16:45:51+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/go-nvml/releases/tag/v0.13.1-0","signal_url":"https://onlylabs.fyi/signals/6cf3ea66-0770-4bfc-8b68-d17429fd2512","signal_json_url":"https://onlylabs.fyi/signals/6cf3ea66-0770-4bfc-8b68-d17429fd2512/signal.json","text":"release · NVIDIA/go-nvml v0.13.1-0 · signal_desk=releases · occurred_at=2026-06-08T16:45:51+00:00 · url=https://github.com/NVIDIA/go-nvml/releases/tag/v0.13.1-0 · raw={\"repo\":\"NVIDIA/go-nvml\"}"},{"ref":"E51","kind":"event","title":"NVIDIA/NVSentinel v1.9.0","date":"2026-06-08T15:58:21+00:00","date_source":"source","source_url":"https://github.com/NVIDIA/NVSentinel/releases/tag/v1.9.0","signal_url":"https://onlylabs.fyi/signals/04610c14-9d56-46fd-9e0f-61a63435b429","signal_json_url":"https://onlylabs.fyi/signals/04610c14-9d56-46fd-9e0f-61a63435b429/signal.json","text":"release · NVIDIA/NVSentinel v1.9.0 · signal_desk=releases · occurred_at=2026-06-08T15:58:21+00:00 · url=https://github.com/NVIDIA/NVSentinel/releases/tag/v1.9.0 · raw={\"repo\":\"NVIDIA/NVSentinel\"}"},{"ref":"E52","kind":"event","title":"How the UK Is Turning Sovereign AI Ambition Into Action With NVIDIA Technologies","date":"2026-06-08T06:00:57+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/uk-sovereign-ai-advancements/","signal_url":"https://onlylabs.fyi/signals/16ea9012-cc25-4ed6-9421-350ca98030ec","signal_json_url":"https://onlylabs.fyi/signals/16ea9012-cc25-4ed6-9421-350ca98030ec/signal.json","text":"post_published · How the UK Is Turning Sovereign AI Ambition Into Action With NVIDIA Technologies · signal_desk=talking · occurred_at=2026-06-08T06:00:57+00:00 · url=https://blogs.nvidia.com/blog/uk-sovereign-ai-advancements/ · data_radar_lanes=Infrastructure · data_radar_terms=infra, infrastructure · data_radar_reason=NVIDIA has a writing signal matching infrastructure. · raw={\"excerpt\":\"A year ago at London Tech Week, NVIDIA founder and CEO Jensen Huang and U.K. Prime Minister Keir Starmer made a declaration: the U.K. would be an AI maker, not an AI taker.  At this year’s event, NVIDIA and its partners are showcasing how that commitment is producing real momentum across the nation’s infrastructure, startups […]\"}"},{"ref":"E53","kind":"event","title":"NVIDIA and LG Group Build an AI Factory to Advance Physical AI, Mobility and AI Infrastructure","date":"2026-06-08T03:00:50+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/nvidia-and-lg-group-ai-factory/","signal_url":"https://onlylabs.fyi/signals/ce9088c8-6402-43b7-9624-07902c99148e","signal_json_url":"https://onlylabs.fyi/signals/ce9088c8-6402-43b7-9624-07902c99148e/signal.json","text":"post_published · NVIDIA and LG Group Build an AI Factory to Advance Physical AI, Mobility and AI Infrastructure · signal_desk=talking · occurred_at=2026-06-08T03:00:50+00:00 · url=https://blogs.nvidia.com/blog/nvidia-and-lg-group-ai-factory/ · data_radar_lanes=Data demand, Infrastructure · data_radar_terms=data, infra, infrastructure, gpu · data_radar_reason=NVIDIA has a writing signal matching data demand, infrastructure. · raw={\"excerpt\":\"NVIDIA and LG Group are building an AI factory to accelerate LG Group’s next wave of AI-driven businesses, spanning robotics, autonomous driving, data center technologies and GPU cloud services. The AI factory will provide LG Group with accelerated computing infrastructure to train, simulate, validate and deploy AI-based applications across its key businesses.  The collaboration brings […]\"}"},{"ref":"E54","kind":"event","title":"NVIDIA and Doosan Group Collaborate to Advance Physical AI and AI Factory Infrastructure","date":"2026-06-07T23:00:36+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/nvidia-and-doosan-group-physical-ai/","signal_url":"https://onlylabs.fyi/signals/3eb8d789-d6f2-4a18-b5b1-f765c24de6c0","signal_json_url":"https://onlylabs.fyi/signals/3eb8d789-d6f2-4a18-b5b1-f765c24de6c0/signal.json","text":"post_published · NVIDIA and Doosan Group Collaborate to Advance Physical AI and AI Factory Infrastructure · signal_desk=talking · occurred_at=2026-06-07T23:00:36+00:00 · url=https://blogs.nvidia.com/blog/nvidia-and-doosan-group-physical-ai/ · data_radar_lanes=Infrastructure · data_radar_terms=infra, infrastructure, platform · data_radar_reason=NVIDIA has a writing signal matching infrastructure. · raw={\"excerpt\":\"NVIDIA and Doosan Group are expanding their collaboration to advance new opportunities across physical AI, robotics and AI factory infrastructure, spanning Doosan Robotics, Doosan Bobcat, Doosan Enerbility and Doosan Corporation Electro-Materials BG. The collaboration will bring together NVIDIA’s full-stack accelerated computing platforms with Doosan Group’s capabilities in industrial automation, power generation and advanced electronics materials […]\"}"},{"ref":"E55","kind":"event","title":"NVIDIA, KRAFTON, NC and Reigning ‘League of Legends’ Champions T1 Celebrate RTX Spark at Korea’s PC Bangs","date":"2026-06-07T07:00:15+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/krafton-nc-t1-korea-gaming-pc-bang-rtx-spark/","signal_url":"https://onlylabs.fyi/signals/1d5f898d-93d2-4c20-b568-097c90fbeacc","signal_json_url":"https://onlylabs.fyi/signals/1d5f898d-93d2-4c20-b568-097c90fbeacc/signal.json","text":"post_published · NVIDIA, KRAFTON, NC and Reigning ‘League of Legends’ Champions T1 Celebrate RTX Spark at Korea’s PC Bangs · signal_desk=talking · occurred_at=2026-06-07T07:00:15+00:00 · url=https://blogs.nvidia.com/blog/krafton-nc-t1-korea-gaming-pc-bang-rtx-spark/ · raw={\"excerpt\":\"At GTC Taipei at COMPUTEX last week, NVIDIA unveiled RTX Spark, the superchip that reinvents Windows PCs for the era of personal AI agents. On the heels of this announcement, NVIDIA founder and CEO Jensen Huang headed to South Korea, where he introduced RTX Spark to the nation’s passionate gaming community. Leading game developers — […]\"}"},{"ref":"E56","kind":"event","title":"Seoul Purpose: How NVIDIA and South Korea Are Building the Future of AI","date":"2026-06-05T05:38:37+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/korea-ecosystem-2026/","signal_url":"https://onlylabs.fyi/signals/5cc20bd3-6a44-4573-988d-fff06b88959f","signal_json_url":"https://onlylabs.fyi/signals/5cc20bd3-6a44-4573-988d-fff06b88959f/signal.json","text":"post_published · Seoul Purpose: How NVIDIA and South Korea Are Building the Future of AI · signal_desk=talking · occurred_at=2026-06-05T05:38:37+00:00 · url=https://blogs.nvidia.com/blog/korea-ecosystem-2026/ · data_radar_lanes=Infrastructure · data_radar_terms=infra, infrastructure · data_radar_reason=NVIDIA has a writing signal matching infrastructure. · raw={\"excerpt\":\"Home to cutting-edge sovereign AI infrastructure and robotics innovators, as well as one of the world’s most passionate gaming communities, South Korea is one of the world’s centers of AI. NVIDIA founder and CEO Jensen Huang is in Seoul this week to meet the partners and builders behind that work. Monday, June 8, 10:00 a.m. […]\"}"},{"ref":"E57","kind":"event","title":"Forecast: Fun Ahead — 18 Games Join in June to Stream on GeForce NOW","date":"2026-06-04T13:00:27+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/geforce-now-thursday-june-2026-games-list/","signal_url":"https://onlylabs.fyi/signals/84a84853-f1e2-4cc8-ab18-f42a711bb63b","signal_json_url":"https://onlylabs.fyi/signals/84a84853-f1e2-4cc8-ab18-f42a711bb63b/signal.json","text":"post_published · Forecast: Fun Ahead — 18 Games Join in June to Stream on GeForce NOW · signal_desk=talking · occurred_at=2026-06-04T13:00:27+00:00 · url=https://blogs.nvidia.com/blog/geforce-now-thursday-june-2026-games-list/ · raw={\"excerpt\":\"June’s forecast with GeForce NOW: 100% chance of gaming. GeForce NOW is lining up new adventures for the month, from big-name blockbusters to quirky indies ready for the spotlight. Members can dive into fresh worlds, squad up in new playlists and discover “just one more run” favorites — all streaming from the cloud, no downloads […]\"}"},{"ref":"E58","kind":"event","title":"NVIDIA Research Unlocks Advanced Grasping, Smarter Autonomous Driving and Agent Training at Scale","date":"2026-06-03T15:00:57+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/cvpr-research-grasping-driving-agent-training/","signal_url":"https://onlylabs.fyi/signals/27509cba-81ee-472d-8b6f-219dfa057024","signal_json_url":"https://onlylabs.fyi/signals/27509cba-81ee-472d-8b6f-219dfa057024/signal.json","text":"post_published · NVIDIA Research Unlocks Advanced Grasping, Smarter Autonomous Driving and Agent Training at Scale · signal_desk=talking · occurred_at=2026-06-03T15:00:57+00:00 · url=https://blogs.nvidia.com/blog/cvpr-research-grasping-driving-agent-training/ · data_radar_lanes=Infrastructure · data_radar_terms=training · data_radar_reason=NVIDIA has a writing signal matching infrastructure. · raw={\"excerpt\":\"What makes a robot gripper useful isn’t that it can pick up one object — it’s that it can pick up the next one, and the one after that, with a tool it’s never held before.  What makes an autonomous vehicle system safe isn’t just that it can reason through a situation — it’s that […]\"}"},{"ref":"E59","kind":"event","title":"NVIDIA Enables the Next Era Of Physical AI Research With Agent Skills For Autonomous Vehicles, Robotics And Vision AI","date":"2026-06-03T15:00:35+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/cvpr-physical-ai-research-agent-skills/","signal_url":"https://onlylabs.fyi/signals/457f4596-92f6-49ad-9faa-cd3a193cd7f6","signal_json_url":"https://onlylabs.fyi/signals/457f4596-92f6-49ad-9faa-cd3a193cd7f6/signal.json","text":"post_published · NVIDIA Enables the Next Era Of Physical AI Research With Agent Skills For Autonomous Vehicles, Robotics And Vision AI · signal_desk=talking · occurred_at=2026-06-03T15:00:35+00:00 · url=https://blogs.nvidia.com/blog/cvpr-physical-ai-research-agent-skills/ · data_radar_lanes=Evals and quality, Infrastructure · data_radar_terms=eval, systems, training · data_radar_reason=NVIDIA has a writing signal matching evals and quality, infrastructure. · raw={\"excerpt\":\"At CVPR, NVIDIA is unveiling new physical AI agent skills that help researchers and developers speed the development of autonomous vehicles, robots and vision AI systems. The core challenge in physical AI research isn’t simply developing stronger models. It’s building a full workflow around them — reconstructing real-world scenes, generating edge-case scenarios, training policies, evaluating […]\"}"},{"ref":"E60","kind":"event","title":"Industrial Software Leaders Build Secure, Autonomous AI Engineers With NVIDIA NemoClaw","date":"2026-06-02T22:00:58+00:00","date_source":"rss.item_date","source_url":"https://blogs.nvidia.com/blog/industrial-software-leaders-secure-autonomous-ai-engineers-nemoclaw/","signal_url":"https://onlylabs.fyi/signals/b2f0abc5-1d72-445a-a789-a1130e664df5","signal_json_url":"https://onlylabs.fyi/signals/b2f0abc5-1d72-445a-a789-a1130e664df5/signal.json","text":"post_published · Industrial Software Leaders Build Secure, Autonomous AI Engineers With NVIDIA NemoClaw · signal_desk=talking · occurred_at=2026-06-02T22:00:58+00:00 · url=https://blogs.nvidia.com/blog/industrial-software-leaders-secure-autonomous-ai-engineers-nemoclaw/ · raw={\"excerpt\":\"Accelerated computing has revolutionized industrial engineering, compressing simulation times from weeks to hours.  Today’s remaining challenges sit in the end-to-end workflow surrounding the simulations: computer-aided design, meshing, simulation setup and debugging, as well as post-processing and generating summary reports of these processes.  At GTC Taipei at COMPUTEX, NVIDIA and more than a dozen engineering software […]\"}"}]}