TL;DR: If your ibm-watsonx-ai Model or ModelInference call fails at construction, check that you passed exactly one of project_id or space_id; one of the two is required, not optional. You can copy the project_id from the Project's Manage tab (Project -> Manage -> General -> Details). Set params as a dict of GenTextParamsMetaNames (call GenParams().get_example_values() to see examples).

## Fix
1. If your ibm-watsonx-ai Model or ModelInference call fails at construction, check that you passed exactly one of project_id or space_id; one of the two is required, not optional.
   Expected: this specific failure stops.
2. You can copy the project_id from the Project's Manage tab (Project -> Manage -> General -> Details).
   Expected: this specific failure stops.
3. Set params as a dict of GenTextParamsMetaNames (call GenParams().get_example_values() to see examples).
   Expected: this specific failure stops.
4. If a valid model id is rejected, the validate=True default may be at fault; pass validate=False for brand-new or custom model ids.
   Expected: this specific failure stops.

## When to use
You hit exactly this: ibm-watsonx-ai: Model needs project_id or space_id, one is required in ibm-watsonx-ai.

## When not to use
A different error, or the same symptom in a different tool. This page only covers the failure above.

## Compatibility
ibm-watsonx-ai.

---

## Vectle citations

- Exact version: <https://vectle.com/skills/skl_Vbh04zWmcl8PihOP1dxp3w/versions/skv_JHeG5v0MIzabpEQejf9nZQ>
- Canonical HTML skill page: <https://vectle.com/skills/skl_Vbh04zWmcl8PihOP1dxp3w>
- Current Markdown URL: <https://vectle.com/skills/skl_Vbh04zWmcl8PihOP1dxp3w/markdown>

## Selected public attribution and original notices/references

```json
{
  "author": "issueatlas",
  "client": null,
  "notices_and_references": [
    "https://ibm.github.io/watsonx-ai-python-sdk/v1.7.1/fm_model.html"
  ]
}
```