Skip to content

Commit 58c11e9

Browse files
committed
docs(installer): record DGX OS Station validation
Signed-off-by: Aaron Erickson <[email protected]>
1 parent 6d35246 commit 58c11e9

6 files changed

Lines changed: 21 additions & 17 deletions

File tree

‎ci/platform-matrix.json‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
{
22
"$comment": "Single source of truth for NemoClaw launch claims and platform support. Covers platforms, inference providers, supported agents, messaging integrations, and deployment paths. Scripts read this to generate README and docs tables. QA/CI update platform/provider rows; the engineering owner reviews other rows. Docs are derived.",
33
"version": "1.1",
4-
"updated": "2026-07-15",
4+
"updated": "2026-07-17",
55

66
"project_status": {
77
"stage": "alpha",
@@ -64,7 +64,7 @@
6464
"status": "deferred",
6565
"prd_priority": "P1",
6666
"ci_tested": false,
67-
"notes": "The PRD marks this platform as P1. Workstation form-factor with NVIDIA GPUs and the same Docker + NVIDIA Container Toolkit + CDI requirements as DGX Spark. Station remains Deferred. On a clean generic Ubuntu 24.04 ARM64 control, the installer detects DGX Station and offers express install with the pinned `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` recipe, including an approximately 352 GB model download, without follow-up provider, model, policy, or sandbox-name choices. No DGX OS or NVIDIA BaseOS release is currently qualified, so those stock images stop before host preparation. DGX OS 7.5 passed host and plain-container GPU checks, managed inference, tool use, and restart/reuse, but CUDA initialization segfaulted inside the real OpenShell sandbox. Track [NVIDIA/OpenShell#2343](https://github.com/NVIDIA/OpenShell/issues/2343) for requalification. Pass `--station-deepseek` to use `deepseek-ai/DeepSeek-V4-Flash` on the generic Ubuntu control while retaining the one-confirmation express flow. The flag requires an interactive terminal, and `/dev/tty` must be available when the installer runs through `curl | bash`. For headless setup on the generic Ubuntu control, select `NEMOCLAW_PROVIDER=install-vllm` and `NEMOCLAW_VLLM_MODEL=deepseek-v4-flash` instead. Direct managed-vLLM onboarding still defaults to `deepseek-ai/DeepSeek-V4-Flash` when no model override is set."
67+
"notes": "The PRD marks this platform as P1. Workstation form-factor with NVIDIA GPUs and the same Docker + NVIDIA Container Toolkit + CDI requirements as DGX Spark. Station remains Deferred. On a Station GB300, the installer accepts either a clean generic Ubuntu 24.04 ARM64 image or 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`. The stock DGX OS path validates the factory GB300 driver and volatile ECC state, Docker and containerd, Buildx, `nvidia-ctk`, CDI, and GPU visibility through both CDI and `--gpus all` before it proceeds; it does not replace host packages or rewrite the runtime. Unknown DGX OS versions, NVIDIA BaseOS images, unsafe release markers, and other Station generations stop before host preparation. The installer offers express install with the pinned `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` recipe, including an approximately 352 GB model download, without follow-up provider, model, policy, or sandbox-name choices. Direct-GPU sandbox policies add `/sys` to the OpenShell filesystem policy as read-only, which lets CUDA inspect GPU topology and device metadata during `cuInit(0)` without granting sysfs write access. 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. Pass `--station-deepseek` to use `deepseek-ai/DeepSeek-V4-Flash` while retaining the one-confirmation express flow. The flag requires an interactive terminal, and `/dev/tty` must be available when the installer runs through `curl | bash`. For headless setup on a qualifying Station, select `NEMOCLAW_PROVIDER=install-vllm` and `NEMOCLAW_VLLM_MODEL=deepseek-v4-flash` instead. Direct managed-vLLM onboarding still defaults to `deepseek-ai/DeepSeek-V4-Flash` when no model override is set."
6868
},
6969
{
7070
"name": "NVIDIA RTX (consumer and Pro workstation GPUs)",
@@ -147,7 +147,7 @@
147147
"name": "Local vLLM (managed install/start)",
148148
"status": "caveated",
149149
"endpoint_type": "Local OpenAI-compatible",
150-
"notes": "Appears by default on DGX Spark and on DGX Station running the generic Ubuntu 24.04 ARM64 control. DGX Station remains Deferred, and no stock DGX OS release is currently qualified. DGX OS 7.5 completed managed-vLLM inference, tool use, and restart/reuse, but CUDA initialization segfaulted inside the real OpenShell sandbox; follow [NVIDIA/OpenShell#2343](https://github.com/NVIDIA/OpenShell/issues/2343). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`. Host must have the NVIDIA Container Toolkit installed and a CDI spec present (`onboard` asserts CDI presence). NemoClaw pins runtime images to immutable digests. DGX Spark and DGX Station models without a model-specific runtime use the `linux/arm64` digest published under `nvcr.io/nvidia/vllm:26.05.post1-py3`; generic Linux NVIDIA GPU hosts use the matching `linux/arm64` or `linux/amd64` digest published under `nvcr.io/nvidia/vllm:26.03.post1-py3`. The DGX Station express installer selects `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` with a pinned Hugging Face revision and the multi-platform index digest published under `vllm/vllm-openai:v0.22.0`; that index resolves to the `linux/arm64` manifest on Station. Direct managed-vLLM profile defaults are listed in `src/lib/inference/vllm-models.ts`: DGX Spark uses `nvidia/Qwen3.6-35B-A3B-NVFP4`, DGX Station uses `deepseek-ai/DeepSeek-V4-Flash`, and Linux NVIDIA GPU uses `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8`. Image pulls from `nvcr.io` require NGC registry login (`docker login nvcr.io`); onboard prompts for the NGC API key when authentication is missing."
150+
"notes": "Appears by default on DGX Spark and qualifying DGX Station GB300 hosts. DGX Station remains Deferred. The Station path accepts a clean generic Ubuntu 24.04 ARM64 image or stock DGX OS `7.2.0`, `7.4.0`, or `7.5.0` after the release marker and factory GPU runtime pass validation. Direct-GPU sandbox policies add `/sys` to the OpenShell filesystem policy as read-only so CUDA initialization can inspect GPU topology and device metadata without granting sysfs write access. 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. Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`. Host must have the NVIDIA Container Toolkit installed and a CDI spec present (`onboard` asserts CDI presence). NemoClaw pins runtime images to immutable digests. DGX Spark and DGX Station models without a model-specific runtime use the `linux/arm64` digest published under `nvcr.io/nvidia/vllm:26.05.post1-py3`; generic Linux NVIDIA GPU hosts use the matching `linux/arm64` or `linux/amd64` digest published under `nvcr.io/nvidia/vllm:26.03.post1-py3`. The DGX Station express installer selects `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` with a pinned Hugging Face revision and the multi-platform index digest published under `vllm/vllm-openai:v0.22.0`; that index resolves to the `linux/arm64` manifest on Station. Direct managed-vLLM profile defaults are listed in `src/lib/inference/vllm-models.ts`: DGX Spark uses `nvidia/Qwen3.6-35B-A3B-NVFP4`, DGX Station uses `deepseek-ai/DeepSeek-V4-Flash`, and Linux NVIDIA GPU uses `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8`. Image pulls from `nvcr.io` require NGC registry login (`docker login nvcr.io`); onboard prompts for the NGC API key when authentication is missing."
151151
}
152152
],
153153

‎docs/get-started/prerequisites.mdx‎

Lines changed: 5 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -67,11 +67,14 @@ It requires a loaded GB300 driver with zero volatile corrected and uncorrected E
6767
This path does not install or replace host packages, enable or restart services, generate CDI configuration, or rewrite the Docker runtime.
6868
The preparation helper writes a private audit log under `~/station-bootstrap-logs`.
6969
Aside from the installer's separately disclosed `docker`-group membership grant when the current trusted account needs Docker access, its only persistent stock-runtime effect is that the digest-pinned visibility-test image can remain in the Docker cache.
70-
These probes establish local container GPU visibility; physical acceptance of the pinned managed-vLLM and Nemotron Ultra recipe remains a separate gate.
70+
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.
71+
Repeat qualification from a clean host and CI coverage remain pending.
72+
For direct GPU sandboxes on both Station images, NemoClaw adds `/sys` to the OpenShell filesystem policy as read-only.
73+
CUDA uses the visible sysfs metadata during `cuInit(0)`, and the policy does not grant sysfs write access.
7174

7275
<Warning title="DGX Station Support Status">
7376
DGX Station remains Deferred.
74-
Product validation of this express-install boundary on physical DGX Station hardware is still pending.
77+
Repeat qualification from a clean host and CI coverage remain pending before any support-status promotion.
7578
</Warning>
7679

7780
<Warning title="Docker Group Access">

‎docs/get-started/quickstart.mdx‎

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -130,8 +130,8 @@ Use these details when your first-run path needs more control.
130130
After it changes pinned packages, the installer exits with status `10` at the required reboot boundary; reboot, sign in, and run the printed exact-commit command to resume the accepted recipe without another prompt.
131131
On stock DGX OS, the installer instead validates the existing GB300 driver and volatile ECC state, local Docker and Buildx, `nvidia-ctk` and CDI, and `nvidia-smi` device visibility through both GPU container launch paths.
132132
It does not install or replace host packages, enable or restart services, or rewrite the Docker runtime; aside from the private preparation log and separately disclosed `docker`-group membership grant when needed, only the digest-pinned visibility-test image can remain cached.
133-
The visibility probes do not replace physical acceptance of the pinned managed-vLLM and Nemotron Ultra recipe.
134-
This automation does not change Station's Deferred support status; product validation on physical DGX Station hardware remains pending.
133+
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.
134+
This automation does not change Station's Deferred support status; repeat qualification from a clean host and CI coverage remain pending.
135135
Pass `--station-deepseek` to use DeepSeek V4 Flash for a Station demo instead.
136136
The flag selects the interactive express prompt and requires terminal access.
137137
Refer to [Platform Support](../reference/platform-support) and [Choose an Inference Provider](../inference/learn-and-choose/choose-inference-provider) for the current platform behavior.
@@ -213,12 +213,12 @@ Use these details when your first-run path needs more control.
213213
Changing pinned packages exits with status `10` for a reboot; after you sign in and run the printed exact-commit command, the accepted express recipe resumes without another prompt.
214214
On stock DGX OS, the path instead validates the existing GB300 driver and zero-error volatile ECC state, local Docker and Buildx, `nvidia-ctk` and CDI, and `nvidia-smi` device visibility through both CDI and `--gpus all`.
215215
It does not install or replace host packages, enable or restart services, generate CDI configuration, or rewrite the Docker runtime; aside from the private preparation log and separately disclosed `docker`-group membership grant when needed, only the digest-pinned visibility-test image can remain cached.
216-
The visibility probes do not replace physical acceptance of the pinned managed-vLLM and Nemotron Ultra recipe.
217-
This automation does not change Station's Deferred support status; product validation remains pending.
216+
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.
217+
This automation does not change Station's Deferred support status; repeat qualification from a clean host and CI coverage remain pending.
218218
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`.
219219
The `--station-deepseek` flag requires an interactive terminal; in a `curl | bash` pipeline, `/dev/tty` must be available.
220220
Without terminal access, the installer stops before it installs Docker or build dependencies instead of ignoring the flag.
221-
For a headless or CI install on a prepared DGX Station running the qualified generic Ubuntu 24.04 ARM64 control, omit the flag and select the same managed-vLLM recipe explicitly.
221+
For a headless or CI install on a qualifying DGX Station GB300 after host preparation, omit the flag and select the same managed-vLLM recipe explicitly.
222222

223223
```bash
224224
curl -fsSL https://www.nvidia.com/nemoclaw.sh | \
@@ -237,7 +237,7 @@ Use these details when your first-run path needs more control.
237237

238238
<Warning>
239239
Express install automates the configuration but does not change DGX Station's Deferred support status.
240-
Product validation of this express-install boundary is still pending.
240+
One physical DGX OS 7.5 GB300 validation completed; repeat qualification from a clean host and CI coverage remain pending.
241241
</Warning>
242242

243243
The installer auto-launches `nemoclaw onboard` when it can find the new binary.

‎docs/inference/choose-inference-provider.mdx‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,7 @@ Use this status table to distinguish validated provider integrations from adapte
3232
| 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). |
3333
| 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`. |
3434
| 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. |
35-
| Local vLLM (managed install/start) | Tested with limitations | Local OpenAI-compatible | Appears by default on DGX Spark and on DGX Station running the generic Ubuntu 24.04 ARM64 control. DGX Station remains Deferred, and no stock DGX OS release is currently qualified. DGX OS 7.5 completed managed-vLLM inference, tool use, and restart/reuse, but CUDA initialization segfaulted inside the real OpenShell sandbox; follow [NVIDIA/OpenShell#2343](https://github.com/NVIDIA/OpenShell/issues/2343). Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`. Host must have the NVIDIA Container Toolkit installed and a CDI spec present (`onboard` asserts CDI presence). NemoClaw pins runtime images to immutable digests. DGX Spark and DGX Station models without a model-specific runtime use the `linux/arm64` digest published under `nvcr.io/nvidia/vllm:26.05.post1-py3`; generic Linux NVIDIA GPU hosts use the matching `linux/arm64` or `linux/amd64` digest published under `nvcr.io/nvidia/vllm:26.03.post1-py3`. The DGX Station express installer selects `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` with a pinned Hugging Face revision and the multi-platform index digest published under `vllm/vllm-openai:v0.22.0`; that index resolves to the `linux/arm64` manifest on Station. Direct managed-vLLM profile defaults are listed in `src/lib/inference/vllm-models.ts`: DGX Spark uses `nvidia/Qwen3.6-35B-A3B-NVFP4`, DGX Station uses `deepseek-ai/DeepSeek-V4-Flash`, and Linux NVIDIA GPU uses `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8`. Image pulls from `nvcr.io` require NGC registry login (`docker login nvcr.io`); onboard prompts for the NGC API key when authentication is missing. |
35+
| 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. The Station path accepts a clean generic Ubuntu 24.04 ARM64 image or stock DGX OS `7.2.0`, `7.4.0`, or `7.5.0` after the release marker and factory GPU runtime pass validation. Direct-GPU sandbox policies add `/sys` to the OpenShell filesystem policy as read-only so CUDA initialization can inspect GPU topology and device metadata without granting sysfs write access. 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. Generic Linux NVIDIA GPU hosts require `NEMOCLAW_EXPERIMENTAL=1` or `NEMOCLAW_PROVIDER=install-vllm`. Host must have the NVIDIA Container Toolkit installed and a CDI spec present (`onboard` asserts CDI presence). NemoClaw pins runtime images to immutable digests. DGX Spark and DGX Station models without a model-specific runtime use the `linux/arm64` digest published under `nvcr.io/nvidia/vllm:26.05.post1-py3`; generic Linux NVIDIA GPU hosts use the matching `linux/arm64` or `linux/amd64` digest published under `nvcr.io/nvidia/vllm:26.03.post1-py3`. The DGX Station express installer selects `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` with a pinned Hugging Face revision and the multi-platform index digest published under `vllm/vllm-openai:v0.22.0`; that index resolves to the `linux/arm64` manifest on Station. Direct managed-vLLM profile defaults are listed in `src/lib/inference/vllm-models.ts`: DGX Spark uses `nvidia/Qwen3.6-35B-A3B-NVFP4`, DGX Station uses `deepseek-ai/DeepSeek-V4-Flash`, and Linux NVIDIA GPU uses `nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8`. Image pulls from `nvcr.io` require NGC registry login (`docker login nvcr.io`); onboard prompts for the NGC API key when authentication is missing. |
3636
{/* provider-status:end */}
3737

3838
## Hosted Providers

0 commit comments

Comments
 (0)