> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ag2.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# autogen.oai.oai_models.completion_usage.CompletionTokensDetails

<h2 id="autogen.oai.oai_models.completion_usage.CompletionTokensDetails" class="doc doc-heading">
  <code class="doc-symbol doc-symbol-heading doc-symbol-class" />

  <span class="doc doc-object-name doc-class-name">CompletionTokensDetails</span>
</h2>

```python theme={null}
CompletionTokensDetails(**data: Any)
```

!!! abstract "Usage Documentation"
[Models](/docs/api-reference/autogen/oai/oai_models/concepts/models)

A base class for creating Pydantic models.<br />

<b>Parameters:</b>

| Name     | Description   |
| -------- | ------------- |
| `**data` | **Type:** Any |

### Class Attributes

<code class="doc-symbol doc-symbol-heading doc-symbol-attribute" />

#### accepted\_prediction\_tokens

<br />

When using Predicted Outputs, the number of tokens in the prediction that
appeared in the completion.

<code class="doc-symbol doc-symbol-heading doc-symbol-attribute" />

#### audio\_tokens

<br />

Audio input tokens generated by the model.

<code class="doc-symbol doc-symbol-heading doc-symbol-attribute" />

#### model\_config

<br />

<br />

<code class="doc-symbol doc-symbol-heading doc-symbol-attribute" />

#### reasoning\_tokens

<br />

Tokens generated by the model for reasoning.

<code class="doc-symbol doc-symbol-heading doc-symbol-attribute" />

#### rejected\_prediction\_tokens

<br />

When using Predicted Outputs, the number of tokens in the prediction that did
not appear in the completion. However, like reasoning tokens, these tokens are
still counted in the total completion tokens for purposes of billing, output,
and context window limits.