<span id="模型目录与设置"></span>

# Model discovery and settings

Query available models, inspect capabilities, and maintain invocation settings.

<span id="选择任务"></span>

## Choose a task

| Task | Page | Description |
| --- | --- | --- |
| List embedding routing models | [List embedding routing models](list-embedding-models.md) | Workspace declarations enter this list only with at least one ONLINE endpoint having a nonempty address. |
| List routed models | [List routed models](list-routing-models.md) | Combines currently readable Workspace declarations with system/deployment models allowed by the permission scope. |
| List workspace embedding models | [List workspace embedding models](list-workspace-embedding-models.md) | Set `BASE_URL` to the product API prefix `https://api.moi.matrixorigin.cn/v5`; use the corresponding product API URL for a dedicated deployment. |
| Model discovery and probe objects | [Model discovery and probe objects](objects.md) | Use BASE_URL, MOI_KEY, and WORKSPACE_ID from the same target environment. BACKEND_ID comes from that environment's backend list. |
| Probe models on a configured embedding backend | [Probe models on a configured embedding backend](probe-embedding-backend-models.md) | Requires model_backend.read on the specified backend. |
| Probe embedding models | [Probe embedding models](probe-embedding-models.md) | Requires regular access to the current workspace. |
| Probe models on a configured LLM backend | [Probe models on a configured LLM backend](probe-llm-backend-models.md) | Requires model_backend.read on the specified backend. |
| Probe LLM models | [Probe LLM models](probe-llm-models.md) | Requires regular access to the current workspace. |
| Configure an LLM backend and endpoint | [Configure an LLM backend and endpoint](setup-llm-backend.md) | Requires Workspace model_backend.create. |

```{toctree}
:hidden:
:maxdepth: 1

List embedding routing models <list-embedding-models>
List routed models <list-routing-models>
List workspace embedding models <list-workspace-embedding-models>
Model discovery and probe objects <objects>
Probe models on a configured embedding backend <probe-embedding-backend-models>
Probe embedding models <probe-embedding-models>
Probe models on a configured LLM backend <probe-llm-backend-models>
Probe LLM models <probe-llm-models>
Configure an LLM backend and endpoint <setup-llm-backend>
```
