Optional
api{process.env.MISTRAL_API_KEY}
Optional
batchBatch size to use when passing multiple documents to generate
Optional
endpointOverride the default endpoint.
Optional
maxThe maximum number of tokens to generate in the completion. The token count of your prompt plus maxTokens cannot exceed the model's context length.
Optional
modelThe name of the model to use.
"codestral-latest"
Optional
randomThe seed to use for random sampling. If set, different calls will generate deterministic results.
Alias for seed
Optional
streamingWhether or not to stream the response.
{false}
Optional
temperatureWhat sampling temperature to use, between 0.0 and 2.0. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
{0.7}
Optional
topPNucleus sampling, where the model considers the results of the tokens with topP
probability mass.
So 0.1 means only the tokens comprising the top 10% probability mass are considered.
Should be between 0 and 1.
{1}
Generated using TypeDoc
The API key to use.