Serve LLMs with vLLM

Agent-ready models

Agent-ready models are tuned for agentic workloads — tool calling, reasoning traces, and long multi-turn sessions. Use this tab to pick a recommended model for your hardware platform and follow the launch guidance below.

Follow the launch recipe below for your platform to set up the recommended model.

Recommendations by hardware platform

Hardware platform	Recommended agent-ready model	MODEL_HANDLE	Recipe
DGX Spark	Agent-ready Qwen3.6-35B-A3B (NVFP4)	`nvidia/Qwen3.6-35B-A3B-NVFP4`	Launch recipe
DGX Station	DeepSeek-V4-Flash	`deepseek-ai/DeepSeek-V4-Flash`	Launch recipe
RTX PRO	Qwen3.6 27B	`nvidia/Qwen3.6-27B-NVFP4`	Launch recipe

Verify your server

After you launch a recipe above, confirm startup using Watch startup in the Instructions tab, then test with Step 5. Test the API.

Next steps

General serving workflow: Docker setup, health checks, and API testing — see the Instructions tab
Scale out: multi-node serving on multi-node capable hardware — see the Multi-node serving tab

Serve LLMs with vLLM

Agent-ready models

Recommendations by hardware platform

Verify your server

Next steps

Resources