Class NvidiaModel
java.lang.Object
uno.anahata.asi.agi.provider.AbstractModel
uno.anahata.asi.openai.compatible.OpenAiCompatibleModel
uno.anahata.asi.nvidia.NvidiaModel
Concrete model implementation for NVIDIA NIM microservices endpoints.
Handles reasoning content extraction for NVIDIA NIM models, auto-detecting
whether thoughts are returned via reasoning_content fields or <think> tags.
- Author:
- anahata
-
Field Summary
Fields inherited from class AbstractModel
defaultTemperature, defaultTopK, defaultTopP, description, enabled, MARKUP_TABLE_HEADER, provider, rawDescription, supportedActions, supportedResponseModalities, tokenizerType -
Constructor Summary
ConstructorsConstructorDescriptionNvidiaModel(NvidiaAiProvider provider, com.fasterxml.jackson.databind.JsonNode node) Constructs a new NvidiaModel instance from a JSON metadata node.NvidiaModel(NvidiaAiProvider provider, String modelId, String displayName) Constructs a new NvidiaModel instance with explicit model ID and display name. -
Method Summary
Modifier and TypeMethodDescriptionprotected voidenrichPayload(com.fasterxml.jackson.databind.node.ObjectNode payload, GenerationRequest request) Hook for subclasses to add or modify payload parameters before sending the request.Methods inherited from class OpenAiCompatibleModel
buildParametersNode, countTokens, countTokens, countTokens, countTokens, createModelMessage, createResponseWithEstimatedUsage, generateContent, generateContentStream, getAvailableServerTools, getDefaultServerTools, getEndpoint, getProvider, getToolDeclarationJson, isSupportsBatchEmbeddings, isSupportsCachedContent, isSupportsContentGeneration, isSupportsEmbeddings, isSupportsFunctionCalling, preparePayloadMethods inherited from class AbstractModel
getModelId, getProviderId, getTokenizerType, hasDiscrepancy, isRegistered, matches, persist, remove, resetFromApi, toMarkupRow, toSafeModelFileName, toString
-
Constructor Details
-
NvidiaModel
Constructs a new NvidiaModel instance from a JSON metadata node.- Parameters:
provider- The parent NVIDIA AI provider.node- The JSON node containing model metadata.
-
NvidiaModel
Constructs a new NvidiaModel instance with explicit model ID and display name.- Parameters:
provider- The parent NVIDIA AI provider.modelId- The unique model ID.displayName- The human-readable display name.
-
-
Method Details
-
enrichPayload
protected void enrichPayload(com.fasterxml.jackson.databind.node.ObjectNode payload, GenerationRequest request) Description copied from class:OpenAiCompatibleModelHook for subclasses to add or modify payload parameters before sending the request.- Overrides:
enrichPayloadin classOpenAiCompatibleModel- Parameters:
payload- The JSON payload being constructed.request- The generation request.
-

