Optional
avgOutput only. Average log probability score of the candidate.
Optional
citationSource attribution of the generated content.
Optional
contentContains the multi-part content of the response.
Optional
finishDescribes the reason the model stopped generating tokens.
Optional
finishOutput only. The reason why the model stopped generating tokens. If empty, the model has not stopped generating the tokens.
Optional
groundingOutput only. Metadata specifies sources used to ground generated content.
Optional
indexOutput only. Index of the candidate.
Optional
logprobsOutput only. Log-likelihood scores for the response tokens and top tokens
Optional
safetyOutput only. List of ratings for the safety of a response candidate. There is at most one rating per category.
Optional
tokenNumber of tokens for this candidate.
A response candidate generated from the model.