> For the complete documentation index, see [llms.txt](https://help.intangible.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.intangible.ai/intangible-mcp-agents/choosing-a-model.md).

# Choosing a model

Which AI models work with Intangible over MCP, and how to pick one for a large build.

The model behind your agent decides how a long build goes. To build a scene, the agent reads Intangible's orientation and places objects in batches. It stops for your check, then reads the scene back. Larger models hold that sequence together better on big scenes.

## Models that work with Intangible

| App                    | Model                                                           |
| ---------------------- | --------------------------------------------------------------- |
| claude.ai, Claude Code | Claude Opus 5.5                                                 |
| ChatGPT                | ChatGPT's default GPT-6 model, including on a Plus plan         |
| Codex                  | Codex's GPT-6 model (OpenAI models only)                        |
| grok.com, Grok Bot     | Grok, including Grok's free plan                                |
| Grok Build             | Grok 4.7, or an open model such as NVIDIA Nemotron 3.5 Super VL |
| Muse app               | Muse Spark                                                      |
| Muse Code              | Muse Wonder (`muse-wonder-1.0`)                                 |
| Gemini app             | Gemini's default model (personal Google accounts)               |
| Gemini CLI             | Gemini 4 Argon, Gemini 3.8 Flash                                |
| Cursor                 | Cursor's default model, in Agent mode                           |
| Perplexity             | Perplexity's default model                                      |

Two larger models should work through the same connections. Both depend on what your account can reach:

* **Muse Wonder Full** in Muse Code. See [Meta Muse](https://help.intangible.ai/intangible-mcp-agents/connect-muse.md).
* **Nemotron 3 Ultra** in Grok Build. See [NVIDIA Nemotron](https://help.intangible.ai/intangible-mcp-agents/nvidia-nemotron.md).

## Picking a model for a large build

For a large scene, pick the largest model your plan offers in that app. If the agent stops partway, reply "continue". Everything placed so far is saved. Or ask for one zone at a time: "Build the stage area first" lands better than "build the whole wedding".

## Related

* [Agents & MCP](https://help.intangible.ai/intangible-mcp-agents/intangible-mcp-agents.md)
* [What your agent can do](https://help.intangible.ai/intangible-mcp-agents/what-your-agent-can-do.md)
* [Troubleshooting and FAQ](https://help.intangible.ai/intangible-mcp-agents/troubleshooting-and-faq.md)

*Last updated 20 October 2026. Model names change often; pick the largest model your plan offers.*


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.intangible.ai/intangible-mcp-agents/choosing-a-model.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
