Class GeminiCandidate
- Namespace
- Uralstech.UGemini.Models.Generation.Candidate
A response candidate generated from the model.
public class GeminiCandidate : IAppendableData<GeminiCandidate>
- Inheritance
-
GeminiCandidate
- Implements
- Extension Methods
Fields
AvgLogprobs
(No description provided)
public float AvgLogprobs
Field Value
CitationMetadata
Citation information for model-generated candidate.
public GeminiCitationMetadata CitationMetadata
Field Value
Remarks
This field may be populated with recitation information for any text included in Content.
These are passages that are "recited" from copyrighted material in the foundational LLM's training data.
Content
Generated content returned from the model.
public GeminiContent Content
Field Value
FinishReason
The reason why the model stopped generating tokens.
public GeminiFinishReason FinishReason
Field Value
GroundingAttributions
Attribution information for sources that contributed to a grounded answer.
public GeminiGroundingAttribution[] GroundingAttributions
Field Value
Remarks
This field is populated for GenerateAnswer calls.
Only available in the beta API.
Index
Index of the candidate in the list of candidates.
public int Index
Field Value
LogprobsResult
Log-likelihood scores for the response tokens and top tokens
public GeminiLogprobsResult LogprobsResult
Field Value
SafetyRatings
List of ratings for the safety of a response candidate There is at most one rating per category.
public GeminiSafetyRating[] SafetyRatings
Field Value
TokenCount
Token count for this candidate.
public int TokenCount
Field Value
Methods
Append(GeminiCandidate)
Appends the data
to the current IAppendableData<T>.
public void Append(GeminiCandidate data)
Parameters
data
GeminiCandidateThe data to append.