---
title: "Secure AI Agents with OpenShell — Agent-ready Models"
canonical: "https://build.nvidia.com/spark/openshell/agent-ready-models.md"
---

# Agent-ready models

Agent-ready models are tuned for **agentic workloads** — tool calling, reasoning traces, and long multi-turn sessions. Pick the recommended model for your hardware platform, serve it with an OpenAI-compatible API (this playbook uses vLLM), then continue with the **Instructions** tab to wire OpenShell inference routing and launch the OpenClaw sandbox.

## Recommendations by hardware platform

| Hardware platform | Recommended agent-ready model | HuggingFace handle | Notes |
| ----------------- | ----------------------------- | ------------------ | ----- |
| **DGX Spark** | Agent-ready Qwen3.6-35B-A3B (NVFP4) | `nvidia/Qwen3.6-35B-A3B-NVFP4` | Recommended for unified-memory agentic serving; launch via [Agent-ready Models](https://build.nvidia.com/playbooks/vllm/agent-ready-models) |
| **DGX Station** | NVIDIA Nemotron 3 Ultra | `nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4` | Recommended for Station; launch via the [DGX Station deployment guide](https://github.com/NVIDIA-NeMo/nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra/StationDeploymentGuide) |

> [!NOTE]
> **DGX Spark:** launch via [Agent-ready Models](https://build.nvidia.com/playbooks/vllm/agent-ready-models). **DGX Station:** use the [Nemotron 3 Ultra Station deployment guide](https://github.com/NVIDIA-NeMo/nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra/StationDeploymentGuide). OpenShell requires the server to listen on all interfaces (`--host 0.0.0.0`) so the gateway container can reach it — keep that flag when you launch.

## Before you serve

1. Complete Docker setup in the **Instructions** tab (Steps 1–2).
2. Export a Hugging Face token if your model requires it:

```bash
export HF_TOKEN=your_actual_token_here
```

3. Launch the recommended recipe for your hardware platform (Spark: [Agent-ready Models](https://build.nvidia.com/playbooks/vllm/agent-ready-models); Station: [Station deployment guide](https://github.com/NVIDIA-NeMo/nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra/StationDeploymentGuide)), ensuring `--host 0.0.0.0` and port `8000`.
4. Confirm the API is up:

```bash
curl -s http://0.0.0.0:8000/v1/models
```

Expected: a JSON `"data"` array that includes your model handle. If the request hangs, the model is still loading — wait for `Application startup complete` in the container logs and retry.

5. Note the exact model `id` from the response. You will use that string for `openshell inference set` and in the OpenClaw onboarding wizard.

## OpenShell-specific requirements

| Requirement | Why it matters |
| ----------- | -------------- |
| `--host 0.0.0.0` | The OpenShell gateway runs inside Docker and cannot reach a server bound only to `localhost`. |
| Host IP in the provider URL | Create the OpenShell provider with `http://{Machine_IP}:8000/v1`, not `localhost` or `127.0.0.1`. |
| Matching model id | The OpenClaw wizard Model ID must match the handle served by vLLM and configured in `openshell inference set`. |

## Next steps

Return to the **Instructions** tab starting at **Step 5. Serve a model with vLLM** (or skip ahead if the server is already running), then create the inference provider and deploy the sandbox.