Generators
Generators use large language models (LLMs) to generate answers based on the prompt. There's a generator for each supported model provider.
Generators are used in query pipelines, after PromptBuilder and before AnswerBuilder. Choose the generator that works with the model you want to use. deepset Cloud offers the following generators:
- AmazonBedrockGenerator: Uses models hosted on Amazon Bedrock.
- AzureOpenAIGenerator: Uses OpenAI LLMs through Azure services.
- CohereGenerator: Uses Cohere models.
- DeepsetAmazonBedrockGenerator: Uses models hosted on Amazon Bedrock through deepset's account, so you don't need to create your own account.
- DeepsetAzureOpenAIVisionGenerator: Uses GPT-3 and GPT-4 models hosted on Azure to generate answers based on images.
- GoogleAIGeminiGenerator: Uses Google's Gemini models to generate text.
- NvidiaGenerator: Uses models form NVIDIA NIM.
- OpenAIGenerator: Uses OpenAI models.
- SagemakerGenerator: Uses models hosted on Amazon SageMaker.
Updated about 2 months ago