OpenLLM

What sampling temperature to use.

The maximum number of tokens to generate in the completion.

Total probability mass of tokens to consider at each step.

Penalizes repeated tokens according to frequency.

Penalizes repeated tokens.

How many completions to generate for each prompt.

Generates best_of completions server-side and returns the "best".

Holds any model parameters valid for create call not explicitly specified.

Automatically inferred from env var OPENAI_API_KEY if not provided.

Base URL path for API requests, leave blank if not using a proxy or service

Automatically inferred from env var OPENAI_ORG_ID if not provided.

Batch size to use when passing multiple documents to generate.

Timeout for requests to OpenAI completion API. Can be float, httpx.Timeout or

Adjust the probability of specific tokens being generated.

Maximum number of retries to make when generating.

Whether to stream the results or not.

Set of special tokens that are allowed。

Set of special tokens that are not allowed。

The model name to pass to tiktoken when using this class.

Optional httpx.Client.

Get max context size for this model.