diff --git a/ci/platform-matrix.json b/ci/platform-matrix.json index c2ad65f40f0..8b4ac2b0f98 100644 --- a/ci/platform-matrix.json +++ b/ci/platform-matrix.json @@ -62,11 +62,11 @@ { "name": "DGX OS (Station)", "runtimes": ["Docker"], - "status": "deferred", + "status": "caveated", "prd_priority": "P1", "ci_tested": false, - "prerequisites_notes": "Deferred. For evaluation on a qualifying DGX Station GB300, see [Additional Setup for DGX Station](additional-setup/dgx-station-preparation) before the Quickstart.", - "notes": "The PRD marks this platform as P1. DGX Station remains Deferred without dedicated CI. Direct-GPU policies expose only the exact read-only BDF directory for each discovered NVIDIA display-class PCI device plus required existing topology and module paths; they do not expose `/sys`, the PCI parent subtree, or sysfs write access. During physical validation, reads of `/sys/fs/cgroup/cgroup.controllers` and `/sys/class/net/lo/address` remained denied. One physical DGX OS `7.5.0` GB300 validation completed with released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. Repeat clean-host qualification and CI coverage remain pending. For canonical hardware qualification, image requirements, preparation, repair limits, reboot handoff, and the explicit temporary metadata override, see [Prepare DGX Station to Install NemoClaw](../get-started/additional-setup/dgx-station-preparation)." + "prerequisites_notes": "Tested with limitations across qualified profiles on one physical DGX Station GB300. Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. See [Additional Setup for DGX Station](additional-setup/dgx-station-preparation) before the Quickstart.", + "notes": "The PRD marks this platform as P1. Physical validation on one DGX Station GB300 covers generic Ubuntu 24.04 ARM64, stock DGX OS `7.5.0`, the April 2026 NVIDIA Colossus BaseOS profile, and the June 2026 NVIDIA AI Developer Tools profile. Clean-host end-to-end validation passed on generic Ubuntu and Colossus BaseOS; stock DGX OS and AI Developer Tools completed Station Express validation. The DGX OS `7.5.0` run used released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. A dual-Station configuration has not been validated, and dedicated CI coverage is not available. Direct-GPU policies expose only the exact read-only BDF directory for each discovered NVIDIA display-class PCI device plus required existing topology and module paths; they do not expose `/sys`, the PCI parent subtree, or sysfs write access. During physical validation, reads of `/sys/fs/cgroup/cgroup.controllers` and `/sys/class/net/lo/address` remained denied. For canonical hardware qualification, image requirements, preparation, repair limits, reboot handoff, and the explicit temporary metadata override, see [Prepare DGX Station to Install NemoClaw](../get-started/additional-setup/dgx-station-preparation)." }, { "name": "NVIDIA RTX (consumer and Pro workstation GPUs)", @@ -149,7 +149,7 @@ "name": "Local vLLM (managed install/start)", "status": "caveated", "endpoint_type": "Local OpenAI-compatible", - "notes": "Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station remains Deferred. For canonical Station qualification and host preparation, see the Additional Setup page for [OpenClaw](/user-guide/openclaw/get-started/additional-setup/dgx-station-preparation), [Hermes](/user-guide/hermes/get-started/additional-setup/dgx-station-preparation), or [Deep Agents](/user-guide/deepagents/get-started/additional-setup/dgx-station-preparation). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`, NVIDIA Container Toolkit, and CDI. NemoClaw pins runtime images to immutable digests. Station Express defaults to `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4`; `--station-deepseek` selects `deepseek-ai/DeepSeek-V4-Flash`. Direct managed-vLLM defaults are `nvidia/Qwen3.6-35B-A3B-NVFP4` on DGX Spark, `deepseek-ai/DeepSeek-V4-Flash` on DGX Station, and `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8` on generic Linux NVIDIA GPU hosts. Image pulls from `nvcr.io` require NGC registry login." + "notes": "Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300; dual-Station configurations are not yet validated, and dedicated CI coverage is not available. For canonical Station qualification and host preparation, see the Additional Setup page for [OpenClaw](/user-guide/openclaw/get-started/additional-setup/dgx-station-preparation), [Hermes](/user-guide/hermes/get-started/additional-setup/dgx-station-preparation), or [Deep Agents](/user-guide/deepagents/get-started/additional-setup/dgx-station-preparation). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`, NVIDIA Container Toolkit, and CDI. NemoClaw pins runtime images to immutable digests. Station Express defaults to `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4`; `--station-deepseek` selects `deepseek-ai/DeepSeek-V4-Flash`. Direct managed-vLLM defaults are `nvidia/Qwen3.6-35B-A3B-NVFP4` on DGX Spark, `deepseek-ai/DeepSeek-V4-Flash` on DGX Station, and `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8` on generic Linux NVIDIA GPU hosts. Image pulls from `nvcr.io` require NGC registry login." } ], diff --git a/docs/changelog/2026-07-14.mdx b/docs/changelog/2026-07-14.mdx index 6f1a7a4f1f8..a8a5d55cc51 100644 --- a/docs/changelog/2026-07-14.mdx +++ b/docs/changelog/2026-07-14.mdx @@ -5,15 +5,18 @@ ## v0.0.83 -NemoClaw v0.0.83 makes shared inference route changes explicit and safe, restores DGX Station GB300 express setup, warns on risky local vLLM configurations, and fixes several onboarding and platform edge cases. +NemoClaw v0.0.83 restores the DGX Station GB300 express-install path without changing its Deferred support status. +It also makes shared inference route changes explicit and safe, warns on risky local vLLM configurations, and fixes several onboarding and platform edge cases. - Shared inference route changes are now explicit and fail-safe. When multiple sandboxes share a gateway, onboarding warns immediately before re-pointing the live route and fails closed before replacing a provider-global identity used by another sandbox. Status output shows each sandbox's recorded route, the live route, and whether `connect` can safely restore a drifted route. For more information, refer to [Use Shared Gateway Routes](/user-guide/openclaw/inference/manage-inference/use-shared-gateway-routes) and [View Active Inference Route](/user-guide/openclaw/inference/manage-inference/view-active-inference-route). -- DGX Station GB300 systems enter the express-install path, and managed vLLM storage preflight treats verified shortages as advisory during express and other non-interactive setup while the checks mature. +- Qualifying DGX Station GB300 systems enter the express-install path. + This path availability does not change DGX Station's Deferred support status. + Managed vLLM storage preflight treats verified shortages as advisory during express and other non-interactive setup while the checks mature. Interactive setup still requires explicit confirmation, and non-interactive setup still stops when model-cache capacity is inconclusive. - For more information, refer to [Set Up vLLM](/user-guide/openclaw/inference/local-inference/set-up-vllm) and [NemoClaw CLI Commands Reference](/user-guide/openclaw/reference/commands). + For more information, refer to [Set Up vLLM](/user-guide/openclaw/inference/local-inference/set-up-vllm), [Platform Support and Launch Claims](/user-guide/openclaw/reference/platform-support), and [NemoClaw CLI Commands Reference](/user-guide/openclaw/reference/commands). - Onboarding warns when a bring-your-own vLLM server on DGX Spark appears to serve a large unquantized model that may exhaust GPU memory under agent tool-call load. The warning is suppressed for the managed Spark vLLM recipe. For more information, refer to [Set Up vLLM](/user-guide/openclaw/inference/local-inference/set-up-vllm) and [Troubleshooting](/user-guide/openclaw/reference/troubleshooting). diff --git a/docs/get-started/dgx-station-preparation.mdx b/docs/get-started/dgx-station-preparation.mdx index 0e4b7942b88..268784dcae6 100644 --- a/docs/get-started/dgx-station-preparation.mdx +++ b/docs/get-started/dgx-station-preparation.mdx @@ -4,7 +4,7 @@ title: "Prepare DGX Station to Install NemoClaw" sidebar-title: "Additional Setup for DGX Station" description: "Qualify and prepare a DGX Station GB300 for NemoClaw express install." -description-agent: "Covers DGX Station GB300 qualification, generic Ubuntu preparation, stock DGX OS validation, and reboot handoff. Use when preparing a DGX Station for NemoClaw express install." +description-agent: "Covers DGX Station GB300 qualification, generic Ubuntu preparation, qualified factory-image validation, and reboot handoff. Use when preparing a DGX Station for NemoClaw express install." keywords: ["nemoclaw dgx station setup", "nemoclaw dgx station gb300", "nemoclaw station express install"] content: type: "reference" @@ -109,7 +109,7 @@ sudo reboot After the host restarts, rerun the NemoClaw installer. If the host uses RDMA, repair OpenIB/OFED instead, then rerun the installer after `openibd.service` is healthy. -## Validate Stock DGX OS +## Validate DGX OS and Factory Images On a qualifying stock DGX OS or exact no-OTA factory image, express install validates the factory stack in place against the local default Docker daemon. It requires a loaded GB300 driver with zero volatile corrected and uncorrected ECC errors, active Docker and containerd services, working Docker Buildx, `nvidia-ctk`, the `nvidia.com/gpu=all` CDI device, and successful `nvidia-smi` device-visibility probes through both CDI and `--gpus all`. @@ -125,13 +125,15 @@ The preparation helper writes a private audit log under `~/station-bootstrap-log Aside from the installer's separately disclosed `docker`-group membership grant when the current trusted account needs Docker access, the stock DGX OS path leaves only the digest-pinned visibility-test image in the Docker cache. The AI Developer Tools path can also leave the packaged CDI refresh units enabled when it repairs a missing device. -One physical DGX OS `7.5.0` GB300 validation completed with released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. -Repeat qualification from a clean host and CI coverage remain pending. +Physical validation on one DGX Station GB300 covers generic Ubuntu 24.04 ARM64, stock DGX OS `7.5.0`, the April 2026 NVIDIA Colossus BaseOS profile, and the June 2026 NVIDIA AI Developer Tools profile. +Clean-host end-to-end validation passed on generic Ubuntu and Colossus BaseOS; stock DGX OS and AI Developer Tools completed Station Express validation. +The DGX OS `7.5.0` run used released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. +A dual-Station configuration has not been validated, and dedicated CI coverage is not available. For the complete support status and direct GPU policy boundaries, see [Platform Support](../../reference/platform-support). -DGX Station remains Deferred. -Repeat qualification from a clean host and CI coverage remain pending before any support-status promotion. +DGX Station is Tested with limitations across the qualified profiles above on one physical DGX Station GB300. +Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. ## Next Step diff --git a/docs/get-started/prerequisites.mdx b/docs/get-started/prerequisites.mdx index c6b77d734c9..25b44e78821 100644 --- a/docs/get-started/prerequisites.mdx +++ b/docs/get-started/prerequisites.mdx @@ -82,15 +82,15 @@ Refer to the [troubleshooting guide](../reference/troubleshooting) for the overr ## Platforms -The following table lists tested platforms and deferred platforms with a documented preparation path. -A Deferred row is available for evaluation only and is not a support claim. +The following table lists platforms with a documented installation path. +A Tested with limitations row applies only to the configuration in its notes. The table comes from [`ci/platform-matrix.json`](https://github.com/NVIDIA/NemoClaw/blob/main/ci/platform-matrix.json), the single source of truth kept in sync by CI and QA. {/* platform-matrix:begin */} | OS | Container runtime | Status | Notes | |----|-------------------|--------|-------| | DGX OS (Spark) | Docker | Tested | Use the standard installer and `$$nemoclaw onboard`. For an end-to-end walkthrough with local inference, see the [NVIDIA Spark playbook](https://build.nvidia.com/spark/nemoclaw). | -| DGX OS (Station) | Docker | Deferred | Deferred. For evaluation on a qualifying DGX Station GB300, see [Additional Setup for DGX Station](additional-setup/dgx-station-preparation) before the Quickstart. | +| DGX OS (Station) | Docker | Tested with limitations | Tested with limitations across qualified profiles on one physical DGX Station GB300. Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. See [Additional Setup for DGX Station](additional-setup/dgx-station-preparation) before the Quickstart. | | Linux | Docker | Tested | Primary tested path. Ubuntu 24.04 has host-level onboarding validation. A digest-pinned Ubuntu 26.04 userspace lane builds the CLI and runs preflight, installer, and platform contracts on eligible main pushes; Docker-host, AppArmor, Landlock, and live onboarding validation on 26.04 remain pending. Other distros (Ubuntu 22.04, Fedora, Rocky, Alma, NixOS, Arch) may work but are not validated. | | macOS (Apple Silicon) | Colima, Docker Desktop | Tested with limitations | Start the container runtime (Colima or Docker Desktop) before running the installer. Homebrew Colima users must install both Colima and the Docker CLI (`brew install colima docker`) before `docker info` can work. Xcode Command Line Tools (`xcode-select --install`) are typically required for Node native modules during install. NemoClaw recommends them but does not enforce them during preflight. | | Windows WSL2 | Docker Desktop (WSL backend) | Tested with limitations | Requires WSL2 with Docker Desktop backend. See [Additional Setup for Windows Machines](additional-setup/windows-preparation) before the Quickstart. | @@ -105,8 +105,9 @@ Use only the page that matches your host. ### DGX Station Express Preparation -DGX Station remains Deferred. -If you are evaluating a qualifying DGX Station GB300, follow [Prepare DGX Station to Install NemoClaw](additional-setup/dgx-station-preparation) before the Quickstart. +DGX Station is Tested with limitations on one qualifying DGX Station GB300. +Dual-Station configurations are not yet validated. +Follow [Prepare DGX Station to Install NemoClaw](additional-setup/dgx-station-preparation) before the Quickstart. ### Windows Preparation diff --git a/docs/get-started/quickstart.mdx b/docs/get-started/quickstart.mdx index 2611adbe20f..44b120c2d64 100644 --- a/docs/get-started/quickstart.mdx +++ b/docs/get-started/quickstart.mdx @@ -121,11 +121,14 @@ Use these details when your first-run path needs more control. DGX Spark, qualifying Station GB300 hosts, and Windows WSL offer an interactive express-install path that chooses a managed local inference option for the platform. Station accepts the generic Ubuntu 24.04 ARM64 image and stock DGX OS `7.2.0`, `7.4.0`, or `7.5.0` when a safe, root-owned `/etc/dgx-release` marker identifies `DGX Server for GALAXY-GB300`. + It also accepts the exact April 2026 NVIDIA Colossus BaseOS and June 2026 NVIDIA AI Developer Tools profiles described in the Station preparation guide. On a qualifying Station, accepting the prompt selects the pinned `nemotron-3-ultra-550b-a55b` managed-vLLM recipe and completes onboarding without more provider, model, policy, or sandbox-name choices. [Prepare DGX Station to Install NemoClaw](additional-setup/dgx-station-preparation) defines Station qualification, generic Ubuntu preparation, stock DGX OS validation, repair limits, and reboot handoff. - By default, unknown versions, unsafe release markers, NVIDIA BaseOS images, and other Station generations stop before host preparation; set `NEMOCLAW_PROVIDER` or `NEMOCLAW_NO_EXPRESS=1` to bypass Station host automation. + By default, unknown versions, unsafe release markers, unmatched no-OTA factory images, and other Station generations stop before host preparation; set `NEMOCLAW_PROVIDER` or `NEMOCLAW_NO_EXPRESS=1` to bypass Station host automation. For an explicit temporary override on genuine Station GB300 hardware with unrecognized release metadata, follow the `--force-station-install` safeguards in the Station preparation guide. - One physical DGX OS `7.5.0` GB300 validation completed, but Station remains Deferred pending repeat clean-host qualification and CI coverage. + Physical single-Station validation covers generic Ubuntu 24.04 ARM64, stock DGX OS `7.5.0`, the April 2026 NVIDIA Colossus BaseOS profile, and the June 2026 NVIDIA AI Developer Tools profile. + Clean-host end-to-end validation passed on generic Ubuntu and Colossus BaseOS; stock DGX OS and AI Developer Tools completed Station Express validation. + Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. Pass `--station-deepseek` to use DeepSeek V4 Flash for a Station demo instead; the flag selects the interactive prompt and requires terminal access. Refer to [Platform Support](../reference/platform-support) and [Choose an Inference Provider](../inference/learn-and-choose/choose-inference-provider) for the current platform behavior. @@ -199,8 +202,9 @@ Use these details when your first-run path needs more control. Before Station host preparation begins, the installer stores the accepted Express recipe in owner-only local state. If preparation requires a reboot or a new login, run the printed command to restore the recorded revision, agent, model, sandbox, policy tier, and gateway, dashboard, and vLLM ports without repeating the Express prompt. Generic Ubuntu preparation can change pinned packages and then exits with status `10`; reboot, sign in, and run that printed command to resume. - Stock DGX OS validation checks the factory stack in place without installing packages, restarting services, or rewriting the Docker runtime. - One physical DGX OS `7.5.0` GB300 validation completed, but Station remains Deferred pending repeat clean-host qualification and CI coverage. + Qualified DGX OS and factory-image validation checks the factory stack in place, with only the bounded repairs documented in the Station preparation guide. + Physical validation on one DGX Station GB300 covers the qualified generic Ubuntu, stock DGX OS `7.5.0`, April 2026 NVIDIA Colossus BaseOS, and June 2026 NVIDIA AI Developer Tools profiles. + Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. To select DeepSeek V4 Flash while retaining the one-confirmation Station express flow, run `curl -fsSL https://www.nvidia.com/nemoclaw.sh | bash -s -- --station-deepseek`. The `--station-deepseek` flag requires an interactive terminal; in a `curl | bash` pipeline, `/dev/tty` must be available. The temporary `--force-station-install` flag has the same terminal requirement and bypasses only DGX release-metadata qualification on genuine Station GB300 hardware. @@ -223,8 +227,8 @@ Use these details when your first-run path needs more control. Set `NEMOCLAW_NO_EXPRESS=1` to skip the express prompt, or set `NEMOCLAW_PROVIDER` before launching the installer to choose a provider yourself. - Express install automates the configuration but does not change DGX Station's Deferred support status. - One physical DGX OS 7.5 GB300 validation completed; repeat qualification from a clean host and CI coverage remain pending. + DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300. + Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. The installer auto-launches `nemoclaw onboard` when it can find the new binary. diff --git a/docs/inference/choose-inference-provider.mdx b/docs/inference/choose-inference-provider.mdx index 660166b4495..6168c05a27d 100644 --- a/docs/inference/choose-inference-provider.mdx +++ b/docs/inference/choose-inference-provider.mdx @@ -30,7 +30,7 @@ Use this status table to distinguish validated provider integrations from adapte | Local Ollama | Tested with limitations | Local Ollama API | Available when Ollama is installed or running on the host. Validated default models: `qwen3.6:35b` (high VRAM), `nemotron-3-nano:30b` (medium VRAM), `qwen3.5:9b` (low VRAM fallback). | | Local NVIDIA NIM | Experimental | Local OpenAI-compatible | Requires `NEMOCLAW_EXPERIMENTAL=1` and a NIM-capable NVIDIA GPU. Host must have the NVIDIA Container Toolkit installed and a CDI spec present (`onboard` asserts CDI presence with `assertCdiNvidiaGpuSpecPresent`, `src/lib/onboard/fatal-runtime-preflight.ts`). NIM images pull from `nvcr.io` and require NGC registry login. NemoClaw gates this path behind the experimental flag because it does not auto-select a NIM image for the host today. You must explicitly pick from the validated image list. On Linux arm64 DGX Spark and DGX Station hosts, onboarding warns that some NIM images may not publish a `linux/arm64` manifest; the warning is advisory, and the selected image pull can still fail when the registry has no matching platform manifest. Managed vLLM has host-specific default models and is not gated on the same boxes. Validated images referenced in `src/lib/inference/config.ts` and `nemoclaw/src/index.ts`: `nvidia/nemotron-3-super-120b-a12b` (default cloud model), `nvidia/nemotron-3-nano-30b-a3b`, `nvidia/llama-3.3-nemotron-super-49b-v1.5`. | | Local vLLM (already running) | Tested with limitations | Local OpenAI-compatible | Appears in the onboarding menu when NemoClaw detects a server already on `localhost:8000`. No flag required. Model is whatever the existing server serves. | -| Local vLLM (managed install/start) | Tested with limitations | Local OpenAI-compatible | Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station remains Deferred. For canonical Station qualification and host preparation, see the Additional Setup page for [OpenClaw](/user-guide/openclaw/get-started/additional-setup/dgx-station-preparation), [Hermes](/user-guide/hermes/get-started/additional-setup/dgx-station-preparation), or [Deep Agents](/user-guide/deepagents/get-started/additional-setup/dgx-station-preparation). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`, NVIDIA Container Toolkit, and CDI. NemoClaw pins runtime images to immutable digests. Station Express defaults to `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4`; `--station-deepseek` selects `deepseek-ai/DeepSeek-V4-Flash`. Direct managed-vLLM defaults are `nvidia/Qwen3.6-35B-A3B-NVFP4` on DGX Spark, `deepseek-ai/DeepSeek-V4-Flash` on DGX Station, and `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8` on generic Linux NVIDIA GPU hosts. Image pulls from `nvcr.io` require NGC registry login. | +| Local vLLM (managed install/start) | Tested with limitations | Local OpenAI-compatible | Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300; dual-Station configurations are not yet validated, and dedicated CI coverage is not available. For canonical Station qualification and host preparation, see the Additional Setup page for [OpenClaw](/user-guide/openclaw/get-started/additional-setup/dgx-station-preparation), [Hermes](/user-guide/hermes/get-started/additional-setup/dgx-station-preparation), or [Deep Agents](/user-guide/deepagents/get-started/additional-setup/dgx-station-preparation). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`, NVIDIA Container Toolkit, and CDI. NemoClaw pins runtime images to immutable digests. Station Express defaults to `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4`; `--station-deepseek` selects `deepseek-ai/DeepSeek-V4-Flash`. Direct managed-vLLM defaults are `nvidia/Qwen3.6-35B-A3B-NVFP4` on DGX Spark, `deepseek-ai/DeepSeek-V4-Flash` on DGX Station, and `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8` on generic Linux NVIDIA GPU hosts. Image pulls from `nvcr.io` require NGC registry login. | {/* provider-status:end */} ## Hosted Providers diff --git a/docs/inference/set-up-vllm.mdx b/docs/inference/set-up-vllm.mdx index 16a7be03448..ee255a379c8 100644 --- a/docs/inference/set-up-vllm.mdx +++ b/docs/inference/set-up-vllm.mdx @@ -163,15 +163,16 @@ NemoClaw intentionally keeps its existing bridge-networked managed-inference top The container publishes port `8000` through Docker, so apply the firewall guidance at the top of this page. -DGX Station remains Deferred. +DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300. On DGX Station GB300, NemoClaw can prepare a clean generic Ubuntu 24.04 ARM64 host for the reviewed Station path. -By default, stock DGX OS `7.2.0`, `7.4.0`, or `7.5.0` requires the supported release marker before NemoClaw validates the factory GPU runtime in place. +Stock DGX OS `7.2.0`, `7.4.0`, or `7.5.0` requires the supported release marker before NemoClaw validates the factory GPU runtime in place; the exact April 2026 NVIDIA Colossus BaseOS and June 2026 NVIDIA AI Developer Tools profiles use their bounded factory-image paths. For the explicit temporary metadata override, follow [Prepare DGX Station to Install NemoClaw](../../get-started/additional-setup/dgx-station-preparation). On Station GB300 only, a direct GPU sandbox receives the exact read-only BDF directory for each host-discovered NVIDIA display-class PCI device, plus the existing CPU, memory, NUMA topology, and NVIDIA module initialization paths. The policy does not grant `/sys`, the PCI parent subtree, or sysfs write access. In the physical DGX OS `7.5.0` validation, this exact-BDF policy passed sandbox `cuInit(0)` while reads of `/sys/fs/cgroup/cgroup.controllers` and `/sys/class/net/lo/address` remained denied. -One physical DGX OS `7.5.0` GB300 validation completed with released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. -DGX Station remains Deferred pending repeat qualification from a clean host and CI coverage. +Physical validation covers generic Ubuntu 24.04 ARM64, stock DGX OS `7.5.0`, the April 2026 NVIDIA Colossus BaseOS profile, and the June 2026 NVIDIA AI Developer Tools profile on one DGX Station GB300. +Clean-host end-to-end validation passed on generic Ubuntu and Colossus BaseOS; stock DGX OS and AI Developer Tools completed Station Express validation. +Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. ## Run Non-Interactive Onboarding diff --git a/docs/reference/platform-support.mdx b/docs/reference/platform-support.mdx index 2afd3914a15..742b3af3493 100644 --- a/docs/reference/platform-support.mdx +++ b/docs/reference/platform-support.mdx @@ -79,7 +79,7 @@ For install requirements and the shorter setup-oriented platform view, refer to | OS | Container runtime | Status | PRD priority | CI | Notes | |----|-------------------|--------|--------------|----|-------| | DGX OS (Spark) | Docker | Tested | P1 | Yes | Use the standard installer and `$$nemoclaw onboard`. For an end-to-end walkthrough with local inference, see the [NVIDIA Spark playbook](https://build.nvidia.com/spark/nemoclaw). | -| DGX OS (Station) | Docker | Deferred | P1 | No | The PRD marks this platform as P1. DGX Station remains Deferred without dedicated CI. Direct-GPU policies expose only the exact read-only BDF directory for each discovered NVIDIA display-class PCI device plus required existing topology and module paths; they do not expose `/sys`, the PCI parent subtree, or sysfs write access. During physical validation, reads of `/sys/fs/cgroup/cgroup.controllers` and `/sys/class/net/lo/address` remained denied. One physical DGX OS `7.5.0` GB300 validation completed with released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. Repeat clean-host qualification and CI coverage remain pending. For canonical hardware qualification, image requirements, preparation, repair limits, reboot handoff, and the explicit temporary metadata override, see [Prepare DGX Station to Install NemoClaw](../get-started/additional-setup/dgx-station-preparation). | +| DGX OS (Station) | Docker | Tested with limitations | P1 | No | The PRD marks this platform as P1. Physical validation on one DGX Station GB300 covers generic Ubuntu 24.04 ARM64, stock DGX OS `7.5.0`, the April 2026 NVIDIA Colossus BaseOS profile, and the June 2026 NVIDIA AI Developer Tools profile. Clean-host end-to-end validation passed on generic Ubuntu and Colossus BaseOS; stock DGX OS and AI Developer Tools completed Station Express validation. The DGX OS `7.5.0` run used released OpenShell `0.0.85`, local Nemotron Ultra serving, sandbox `cuInit(0)`, and a Hermes write/read file-tool task. A dual-Station configuration has not been validated, and dedicated CI coverage is not available. Direct-GPU policies expose only the exact read-only BDF directory for each discovered NVIDIA display-class PCI device plus required existing topology and module paths; they do not expose `/sys`, the PCI parent subtree, or sysfs write access. During physical validation, reads of `/sys/fs/cgroup/cgroup.controllers` and `/sys/class/net/lo/address` remained denied. For canonical hardware qualification, image requirements, preparation, repair limits, reboot handoff, and the explicit temporary metadata override, see [Prepare DGX Station to Install NemoClaw](../get-started/additional-setup/dgx-station-preparation). | | Linux | Docker | Tested | P0 | Yes | Primary tested path. Ubuntu 24.04 has host-level onboarding validation. A digest-pinned Ubuntu 26.04 userspace lane builds the CLI and runs preflight, installer, and platform contracts on eligible main pushes; Docker-host, AppArmor, Landlock, and live onboarding validation on 26.04 remain pending. Other distros (Ubuntu 22.04, Fedora, Rocky, Alma, NixOS, Arch) may work but are not validated. | | macOS (Apple Silicon) | Colima, Docker Desktop | Tested with limitations | P0 | Yes | Start the container runtime (Colima or Docker Desktop) before running the installer. Homebrew Colima users must install both Colima and the Docker CLI (`brew install colima docker`) before `docker info` can work. Xcode Command Line Tools (`xcode-select --install`) are typically required for Node native modules during install. NemoClaw recommends them but does not enforce them during preflight. | | NVIDIA RTX (consumer and Pro workstation GPUs) | Docker | Deferred | P1 | No | The PRD marks this platform as P1. Covers RTX consumer cards and RTX Pro workstation cards on Linux hosts that meet the generic-Linux-GPU requirements (NVIDIA Container Toolkit + CDI present). The provider menu emits managed vLLM behind `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm` for this host class today; the end-to-end onboard path on this hardware is not yet validated in CI. | @@ -104,7 +104,7 @@ NemoClaw routes inference through the OpenShell gateway. Each row below is a pro | Local Ollama | Tested with limitations | Local Ollama API | Available when Ollama is installed or running on the host. Validated default models: `qwen3.6:35b` (high VRAM), `nemotron-3-nano:30b` (medium VRAM), `qwen3.5:9b` (low VRAM fallback). | | Local NVIDIA NIM | Experimental | Local OpenAI-compatible | Requires `NEMOCLAW_EXPERIMENTAL=1` and a NIM-capable NVIDIA GPU. Host must have the NVIDIA Container Toolkit installed and a CDI spec present (`onboard` asserts CDI presence with `assertCdiNvidiaGpuSpecPresent`, `src/lib/onboard/fatal-runtime-preflight.ts`). NIM images pull from `nvcr.io` and require NGC registry login. NemoClaw gates this path behind the experimental flag because it does not auto-select a NIM image for the host today. You must explicitly pick from the validated image list. On Linux arm64 DGX Spark and DGX Station hosts, onboarding warns that some NIM images may not publish a `linux/arm64` manifest; the warning is advisory, and the selected image pull can still fail when the registry has no matching platform manifest. Managed vLLM has host-specific default models and is not gated on the same boxes. Validated images referenced in `src/lib/inference/config.ts` and `nemoclaw/src/index.ts`: `nvidia/nemotron-3-super-120b-a12b` (default cloud model), `nvidia/nemotron-3-nano-30b-a3b`, `nvidia/llama-3.3-nemotron-super-49b-v1.5`. | | Local vLLM (already running) | Tested with limitations | Local OpenAI-compatible | Appears in the onboarding menu when NemoClaw detects a server already on `localhost:8000`. No flag required. Model is whatever the existing server serves. | -| Local vLLM (managed install/start) | Tested with limitations | Local OpenAI-compatible | Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station remains Deferred. For canonical Station qualification and host preparation, see the Additional Setup page for [OpenClaw](/user-guide/openclaw/get-started/additional-setup/dgx-station-preparation), [Hermes](/user-guide/hermes/get-started/additional-setup/dgx-station-preparation), or [Deep Agents](/user-guide/deepagents/get-started/additional-setup/dgx-station-preparation). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`, NVIDIA Container Toolkit, and CDI. NemoClaw pins runtime images to immutable digests. Station Express defaults to `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4`; `--station-deepseek` selects `deepseek-ai/DeepSeek-V4-Flash`. Direct managed-vLLM defaults are `nvidia/Qwen3.6-35B-A3B-NVFP4` on DGX Spark, `deepseek-ai/DeepSeek-V4-Flash` on DGX Station, and `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8` on generic Linux NVIDIA GPU hosts. Image pulls from `nvcr.io` require NGC registry login. | +| Local vLLM (managed install/start) | Tested with limitations | Local OpenAI-compatible | Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300; dual-Station configurations are not yet validated, and dedicated CI coverage is not available. For canonical Station qualification and host preparation, see the Additional Setup page for [OpenClaw](/user-guide/openclaw/get-started/additional-setup/dgx-station-preparation), [Hermes](/user-guide/hermes/get-started/additional-setup/dgx-station-preparation), or [Deep Agents](/user-guide/deepagents/get-started/additional-setup/dgx-station-preparation). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`, NVIDIA Container Toolkit, and CDI. NemoClaw pins runtime images to immutable digests. Station Express defaults to `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4`; `--station-deepseek` selects `deepseek-ai/DeepSeek-V4-Flash`. Direct managed-vLLM defaults are `nvidia/Qwen3.6-35B-A3B-NVFP4` on DGX Spark, `deepseek-ai/DeepSeek-V4-Flash` on DGX Station, and `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8` on generic Linux NVIDIA GPU hosts. Image pulls from `nvcr.io` require NGC registry login. | {/* provider-status-full:end */} ## Messaging Integrations diff --git a/docs/resources/prompt-assets/dgx-station.md b/docs/resources/prompt-assets/dgx-station.md index 7c4015a0d58..5f8a6cbfdae 100644 --- a/docs/resources/prompt-assets/dgx-station.md +++ b/docs/resources/prompt-assets/dgx-station.md @@ -23,7 +23,8 @@ Before asking for consent, explain all of these boundaries: - Official Station preparation may add the trusted local account to the `docker` group, which grants root-equivalent control and is suitable only for a trusted single-user development host. - Official Station preparation may require an operator-controlled reboot and resumes only with the accepted NemoClaw revision. - Nemotron Ultra Express discloses an approximately `352 GB` model download. DeepSeek Express downloads its pinned vLLM container and model data. Both require enough space on the model-cache filesystem and Docker storage. -- DGX Station remains an evaluation path with deferred end-to-end validation on physical hardware, so startup may still fail after readiness checks. +- DGX Station is tested with limitations across qualified profiles on one physical DGX Station GB300. +- Dual-Station configurations are not yet validated, and dedicated CI coverage is not available. Ask: "Which DGX Station Express model would you like?" Choices: diff --git a/docs/resources/starter-prompt.md b/docs/resources/starter-prompt.md index 6311fbb9f2a..b1f327c3715 100644 --- a/docs/resources/starter-prompt.md +++ b/docs/resources/starter-prompt.md @@ -68,9 +68,9 @@ Use `NEMOCLAW_AGENT=langchain-deepagents-code` or `nemo-deepagents onboard` for After the readiness check, load exactly one matching instruction asset before provider selection: -- Confirmed DGX Spark: [DGX Spark Express instructions](https://raw.githubusercontent.com/NVIDIA/NemoClaw/c718a78c5794574a98fdd885d94466c3b6794153/docs/resources/prompt-assets/dgx-spark.md). -- Confirmed DGX Station: [DGX Station installation instructions](https://raw.githubusercontent.com/NVIDIA/NemoClaw/c718a78c5794574a98fdd885d94466c3b6794153/docs/resources/prompt-assets/dgx-station.md). -- Officially detected Windows WSL: [Windows WSL Express instructions](https://raw.githubusercontent.com/NVIDIA/NemoClaw/c718a78c5794574a98fdd885d94466c3b6794153/docs/resources/prompt-assets/windows-wsl.md). +- Confirmed DGX Spark: [DGX Spark Express instructions](https://raw.githubusercontent.com/NVIDIA/NemoClaw/ebf29b024c7a24e69ab8386b68e426d31a4c6821/docs/resources/prompt-assets/dgx-spark.md). +- Confirmed DGX Station: [DGX Station installation instructions](https://raw.githubusercontent.com/NVIDIA/NemoClaw/ebf29b024c7a24e69ab8386b68e426d31a4c6821/docs/resources/prompt-assets/dgx-station.md). +- Officially detected Windows WSL: [Windows WSL Express instructions](https://raw.githubusercontent.com/NVIDIA/NemoClaw/ebf29b024c7a24e69ab8386b68e426d31a4c6821/docs/resources/prompt-assets/windows-wsl.md). Read the matching raw Markdown file completely and follow it before continuing. Do not load a platform asset for any other computer. diff --git a/scripts/install.sh b/scripts/install.sh index 4269fbf6dc1..03f97eac03d 100755 --- a/scripts/install.sh +++ b/scripts/install.sh @@ -3764,7 +3764,7 @@ describe_express_install() { ;; esac printf " Host setup may add this trusted local account to the docker group, which grants root-equivalent control. This flow is only for trusted single-user development hosts; shared or managed hosts require an organization-approved Docker access path.\n" - printf " DGX Station remains Deferred; one DGX OS 7.5 GB300 physical validation passed, with repeat clean-host qualification and CI coverage still pending.\n" + printf " DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300; dual-Station configurations are not yet validated, and dedicated CI coverage is not available.\n" sandbox_summary="${NEMOCLAW_SANDBOX_NAME:-my-assistant}" ;; "Windows WSL") diff --git a/test/install-express-prompt.test.ts b/test/install-express-prompt.test.ts index d1c95a1097f..095de68c1af 100644 --- a/test/install-express-prompt.test.ts +++ b/test/install-express-prompt.test.ts @@ -373,7 +373,10 @@ detect_express_platform expect(output).toMatch( /installs missing pinned driver, Docker, and NVIDIA Container Toolkit packages/, ); - expect(output).toMatch(/DGX Station remains Deferred/); + expect(output).toMatch( + /DGX Station is Tested with limitations across qualified profiles on one physical DGX Station GB300/, + ); + expect(output).toMatch(/dual-Station configurations are not yet validated/); expect(output).toMatch(/Using express install for DGX Station/); expect(output).toMatch( /RESULT NON_INTERACTIVE=1 SUDO_MODE=prompt PROVIDER=install-vllm MODEL=nvidia\/nemotron-3-ultra-550b-a55b VLLM_MODEL=nemotron-3-ultra-550b-a55b POLICY=suggested YES=1 SANDBOX=my-assistant/, diff --git a/test/starter-prompt-docs.test.ts b/test/starter-prompt-docs.test.ts index 8f064b43de6..90eff8363ef 100644 --- a/test/starter-prompt-docs.test.ts +++ b/test/starter-prompt-docs.test.ts @@ -30,7 +30,7 @@ const repoRoot = path.resolve(__dirname, ".."); const starterPromptMarkdownSource = path.join(repoRoot, "docs", "resources", "starter-prompt.md"); // CI resolves this Git commit and byte-compares its prompt-asset blobs with // the local files. The digests independently assert those same immutable bytes. -const promptAssetRevision = "c718a78c5794574a98fdd885d94466c3b6794153"; +const promptAssetRevision = "ebf29b024c7a24e69ab8386b68e426d31a4c6821"; type PromptAsset = { path: string; @@ -53,7 +53,7 @@ const promptAssets = { ), dgxStation: definePromptAsset( "docs/resources/prompt-assets/dgx-station.md", - "f0c61cef93da203cecda2424eb1fc5680d56ffd679a518bfc98d26b2e82be381", // gitleaks:allow -- pinned prompt-asset SHA-256 + "9f506ece27dcda3cf85735d7c6a80846a7727696b825cf8ca161334ac6925c1f", // gitleaks:allow -- pinned prompt-asset SHA-256 ), windowsWsl: definePromptAsset( "docs/resources/prompt-assets/windows-wsl.md", @@ -765,6 +765,13 @@ describe("starter prompt docs CTA", () => { expect(stationSource).not.toContain("provider-preseeded DeepSeek path"); expect(stationSource).not.toContain("only supported next step"); expect(stationSource).toContain("model-cache filesystem and Docker storage"); + expect(stationSource).toContain( + "DGX Station is tested with limitations across qualified profiles on one physical DGX Station GB300.", + ); + expect(stationSource).toContain( + "Dual-Station configurations are not yet validated, and dedicated CI coverage is not available.", + ); + expect(stationSource).not.toContain("deferred end-to-end validation on physical hardware"); expect(stationSource).toContain( "Do not run `scripts/prepare-dgx-station-host.sh --check`, `--verify`, or `--apply` separately", ); @@ -786,8 +793,6 @@ describe("starter prompt docs CTA", () => { ); expect(stationSource).toContain("Run the installer only in a secure interactive terminal"); expect(stationSource).toContain("Keep each official confirmation visible"); - expect(stationSource).toContain("evaluation path with deferred end-to-end validation"); - expect(stationSource).toContain("startup may still fail after readiness checks"); expect(stationSource).toContain("third-party-software notice"); expect(windowsSource).toContain("NEMOCLAW_PROVIDER=install-windows-ollama"); expect(windowsSource).toContain("Do not start a second Ollama service on the same port."); diff --git a/test/station-doc-ownership.test.ts b/test/station-doc-ownership.test.ts index b33de3509e8..07456a5fb32 100644 --- a/test/station-doc-ownership.test.ts +++ b/test/station-doc-ownership.test.ts @@ -62,11 +62,20 @@ describe("DGX Station documentation ownership", () => { expect(platformSupport).toContain("`/sys/fs/cgroup/cgroup.controllers`"); expect(platformSupport).toContain("`/sys/class/net/lo/address`"); expect(vllmSetup).toContain("explicit temporary metadata override"); - expect(stationPreparation).toMatch(/(?:DGX )?Station(?: remains|'s) Deferred/); - expect(stationPreparation).toContain("One physical DGX OS `7.5.0` GB300 validation completed"); + expect(stationPreparation).toContain("DGX Station is Tested with limitations"); + expect(stationPreparation).toContain("Dual-Station configurations are not yet validated"); + expect(stationPreparation).toContain("dedicated CI coverage is not available"); + expect(stationPreparation).toContain( + "Physical validation on one DGX Station GB300 covers generic Ubuntu 24.04 ARM64", + ); + expect(stationPreparation).toContain("April 2026 NVIDIA Colossus BaseOS"); + expect(stationPreparation).toContain("June 2026 NVIDIA AI Developer Tools"); + expect(stationPreparation).toContain( + "Clean-host end-to-end validation passed on generic Ubuntu and Colossus BaseOS", + ); expect(stationPreparation).toContain("[Platform Support](../../reference/platform-support)"); expect(prerequisites).toContain("### DGX Station Express Preparation"); - expect(prerequisites).toMatch(/\| DGX OS \(Station\) \| Docker \| Deferred \|/); + expect(prerequisites).toMatch(/\| DGX OS \(Station\) \| Docker \| Tested with limitations \|/); expect(prerequisites).toContain("additional-setup/dgx-station-preparation"); expect(prerequisites).toContain( "[Additional Setup for DGX Station](additional-setup/dgx-station-preparation)", @@ -76,8 +85,12 @@ describe("DGX Station documentation ownership", () => { ); expect(quickstart).toContain("additional-setup/dgx-station-preparation"); expect(quickstart).not.toContain("prerequisites#dgx-station-express-preparation"); - expect(quickstart).toMatch(/(?:DGX )?Station(?: remains|'s) Deferred/); - expect(quickstart).toContain("One physical DGX OS `7.5.0` GB300 validation completed"); + expect(quickstart).toContain("DGX Station is Tested with limitations"); + expect(quickstart).toContain("Dual-Station configurations are not yet validated"); + expect(quickstart).toContain("dedicated CI coverage is not available"); + expect(quickstart).toContain("unmatched no-OTA factory images"); + expect(quickstart).toContain("April 2026 NVIDIA Colossus BaseOS"); + expect(quickstart).toContain("June 2026 NVIDIA AI Developer Tools"); }); it("labels platform-specific prerequisite pages as additional setup", () => {