Models
IBM WatsonX
The IBM WatsonX model provides access to IBM’s language models.
Parameters
Parameter | Type | Default | Description |
---|---|---|---|
id | str | "ibm/granite-20b-code-instruct" | The model ID to use |
frequency_penalty | float | None | Penalty for using frequent tokens. Higher values discourage repetition |
presence_penalty | float | None | Penalty for using tokens already present in the text. Higher values encourage new topics |
max_tokens | int | None | Maximum number of tokens to generate in the response |
temperature | float | None | Controls randomness in responses. Higher values make output more random |
top_p | float | None | Controls diversity of responses via nucleus sampling |
logprobs | int | None | Number of log probabilities to return |
top_logprobs | int | None | Number of most likely tokens to return log probabilities for |
response_format | Any | None | Format specification for the response |
api_key | str | None | IBM WatsonX API key |
project_id | str | None | IBM WatsonX project ID |
url | str | "https://eu-de.ml.cloud.ibm.com" | IBM WatsonX API endpoint URL |
verify | bool | True | Whether to verify SSL certificates |
client_params | Dict[str, Any] | None | Additional parameters to pass to the client |