# tinker_cookbook.completers.TinkerTokenCompleter

## *class* [**tinker_cookbook.completers.TinkerTokenCompleter**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L88)(*[TokenCompleter](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/completers/tokencompleter/index.md)*)

Token completer that uses a tinker.SamplingClient to sample actions.

```python
completer = TinkerTokenCompleter(sampling_client, max_tokens=512)
result = await completer(model_input, stop=["<|endoftext|>"])
print(result.tokens, result.logprobs)
```

**Fields:**

- [**sampling_client**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L108) (*tinker.[SamplingClient](https://tinker-docs.thinkingmachines.ai/tinker/api-reference/samplingclient/index.md)*)
- [**max_tokens**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L109) (*int*)
- [**temperature**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L110) (*float*, default: `1.0`)
- [**context_window**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L111) (*int | None*, default: `None`)

### [**__call__**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L113)(*model_input*, *stop*, *max_tokens*)

Sample an action from the policy given an observation.

A per-call `max_tokens` override is combined with the completer's own `max_tokens` by taking the minimum, and the `context_window` cap (when configured) still applies on top.

**Parameters:**

- [**model_input**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L115) (*tinker.[ModelInput](https://tinker-docs.thinkingmachines.ai/tinker/api-reference/types/modelinput/index.md)*)
- [**stop**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L116) (*StopCondition*)
- [**max_tokens**](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/completers.py#L118) (*int | None*)

**Returns:** *[TokensWithLogprobs](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/completers/tokenswithlogprobs/index.md)*