Supported Models

This page provides a list of all supported language models and embedding models.

Language Models (LMs)

LMs are used to generate natural language responses, either directly from user prompts or based on retrieved context in a RAG pipeline. Use one of the model IDs below to configure the language model component.

1. Anthropic

   1. Model ID Format: anthropic/model-name.

   2. Example: anthropic/claude-3-5-sonnet-latest.

   3. Anthropic API key: Create one from Anthropic Console.

   4. Available models: Anthropic available models.

2. Azure OpenAI

   1. Model ID Format: azure-openai/azure-endpoint:azure-deployment.

   2. Example: azure-openai/https://my-resource.openai.azure.com:my-deployment.

   3. Azure OpenAI Resource and Deployment:

      1. Create an Azure OpenAI resource.

      2. Deploy a model.

   4. Available models: Azure OpenAI available models.

3. Bedrock

   1. Model ID Format: bedrock/model-name.

   2. Example: bedrock/us.anthropic.claude-sonnet-4-20250514-v1:0.

   3. Bedrock access key id and secret access key:

      1. Create an AWS account and an IAM user.

      2. Create new access keys for an IAM user.

   4. Available models: Bedrock supported models.

4. Google Gen AI

   1. Model ID Format: google/model-name.

   2. Example: google/gemini-1.5-flash-latest.

   3. Gemini API key: Create one from Google AI Studio.

   4. Available models: Google Gen AI available models.

5. Google Vertex AI

   1. Model ID Format: google/model-name.

   2. Example: google/gemini-1.5-flash.

   3. Google Service Account JSON Credential Path:

      1. Create a Google service account.

      2. Create JSON credentials for the service account.

   4. Available models: Google Vertex AI available models.

6. LangChain

   1. Model ID Format: langchain/<package>.<class>:model-name.

   2. Example: langchain/langchain_openai.ChatOpenAI:gpt-4o-mini.

   3. Varied credentials depending on the package and class.

   4. Available models: LangChain supported providers.

7. LiteLLM

   1. Model ID Format: litellm/provider/model-name.

   2. Example: litellm/openai/gpt-4o-mini.

   3. Varied credentials depending on the package and class.

   4. Available models: LiteLLM supported providers.

8. OpenAI

   1. Model ID Format: openai/model-name.

   2. Example: openai/gpt-4o-mini.

   3. Open AI API key: Create one from OpenAI Console.

   4. Available models: OpenAI available models.

9. OpenAI Compatible

   1. Model ID Format: openai-compatible/base-url:model-name.

   2. Example: openai-compatible/https://api.groq.com/openai/v1:llama3-8b-8192.

   3. Varied credentials depending on the endpoints.

   4. Compatible endpoints include but are not limited to:

      1. DeepInfra

      2. DeepSeek

      3. Groq

      4. OpenRouter

      5. Text Generation Inference

      6. Together.ai

      7. vLLM

Embedding Models (EMs)

Embedding models are used to convert text into vector representations, enabling similarity search and document retrieval. Use one of the model IDs below to configure the embedding model component.

1. Azure OpenAI

   1. Model ID Format: azure-openai/azure-endpoint:azure-deployment.

   2. Example: azure-openai/https://my-resource.openai.azure.com:my-deployment.

   3. Azure OpenAI Resource and Deployment:

      1. Create an Azure OpenAI resource.

      2. Deploy a model.

   4. Available models: Azure OpenAI available models.

2. Google Gen AI

   1. Model ID Format: google/model-name.

   2. Example: google/text-embedding-004.

   3. Gemini API key: Create one from Google AI Studio.

   4. Available models: Google Gen AI available models.

3. Google Vertex AI

   1. Model ID Format: google/model-name.

   2. Example: google/text-embedding-004.

   3. Google Service Account JSON Credential Path:

      1. Create a Google service account.

      2. Create JSON credentials for the service account.

   4. Available models: Google Vertex AI available models.

4. LangChain

   1. Model ID Format: langchain/<package>.<class>:model-name.

   2. Example: langchain/langchain_openai.OpenAIEmbeddings:text-embedding-3-small.

   3. Varied credentials depending on the package and class.

   4. Available models: LangChain supported providers.

5. OpenAI

   1. Model ID Format: openai/model-name.

   2. Example: openai/text-embedding-3-small.

   3. Open AI API key: Create one from OpenAI Console.

   4. Available models: OpenAI available models.

6. Voyage

   1. Model ID Format: voyage/model-name.

   2. Example: voyage/voyage-3.5-lite.

   3. Voyage API key: Create one from Voyage Dashboard.

   4. Available models: Voyage available models.