Chat with LLMs Using Open WebUI and Ollama

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, then pull it through Open WebUI’s integrated Ollama stack using either instructions tab.

Recommendations by hardware platform

DGX Spark: Agent-ready Qwen3.6-35B-A3B (Ollama)

Complete the setup through Step 5 in the Open WebUI Remotely tab (through creating the administrator account), or through Step 4 in the Open WebUI on Desktop tab, so Open WebUI is running and you are signed in.

Then pull and select the recommended model in the Open WebUI interface:

1. Click Select a model in the top left corner.
2. Type qwen3.6:35b-a3b in the search field.
3. Click the Pull "qwen3.6:35b-a3b" from Ollama.com button that appears.
4. Wait for the download to finish, then select qwen3.6:35b-a3b from the model dropdown.

Verify in chat

In the chat text area, enter a prompt such as Write me a haiku about GPUs and AI, then press Enter and wait for the response.

Context window guidance

For agentic use, prefer a context window of at least 32K tokens. If your hardware platform has additional memory headroom, 64K or higher is recommended so multi-turn sessions and tools have room to work. Adjust context settings in Open WebUI or the model controls for your session when available.

Next steps

• Remote workflow: see the Open WebUI Remotely tab
• Desktop workflow: see the Open WebUI on Desktop tab
• Browse other models: Ollama library