Embedding Model Configuration

DeepSearcher supports various embedding models to convert text into vector representations for semantic search.

📝 Basic Configuration

config.set_provider_config("embedding", "(EmbeddingModelName)", "(Arguments dict)")

📋 Available Embedding Providers

Provider	Description	Key Features
OpenAIEmbedding	OpenAI's text embedding models	High quality, production-ready
MilvusEmbedding	Built-in embedding models via Pymilvus	Multiple model options
VoyageEmbedding	VoyageAI embedding models	Specialized for search
BedrockEmbedding	Amazon Bedrock embedding	AWS integration
GeminiEmbedding	Google's Gemini embedding	High performance
GLMEmbedding	ChatGLM embeddings	Chinese language support
OllamaEmbedding	Local embedding with Ollama	Self-hosted option
PPIOEmbedding	PPIO cloud embedding	Scalable solution
SiliconflowEmbedding	Siliconflow's models	Enterprise support
VolcengineEmbedding	Volcengine embedding	High throughput
NovitaEmbedding	Novita AI embedding	Cost-effective

🔍 Provider Examples

OpenAI Embedding

config.set_provider_config("embedding", "OpenAIEmbedding", {"model": "text-embedding-3-small"})

Requires OPENAI_API_KEY environment variable

Milvus Built-in Embedding

config.set_provider_config("embedding", "MilvusEmbedding", {"model": "BAAI/bge-base-en-v1.5"})

config.set_provider_config("embedding", "MilvusEmbedding", {"model": "jina-embeddings-v3"})

For Jina's embedding model, requires JINAAI_API_KEY environment variable

VoyageAI Embedding

config.set_provider_config("embedding", "VoyageEmbedding", {"model": "voyage-3"})

Requires VOYAGE_API_KEY environment variable and pip install voyageai

📚 Additional Providers

Amazon Bedrock

config.set_provider_config("embedding", "BedrockEmbedding", {"model": "amazon.titan-embed-text-v2:0"})

Requires AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY environment variables and pip install boto3

Novita AI

config.set_provider_config("embedding", "NovitaEmbedding", {"model": "baai/bge-m3"})

Requires NOVITA_API_KEY environment variable

Siliconflow

config.set_provider_config("embedding", "SiliconflowEmbedding", {"model": "BAAI/bge-m3"})

Requires SILICONFLOW_API_KEY environment variable

Volcengine

config.set_provider_config("embedding", "VolcengineEmbedding", {"model": "doubao-embedding-text-240515"})

Requires VOLCENGINE_API_KEY environment variable

GLM

config.set_provider_config("embedding", "GLMEmbedding", {"model": "embedding-3"})

Requires GLM_API_KEY environment variable and pip install zhipuai

Google Gemini

config.set_provider_config("embedding", "GeminiEmbedding", {"model": "text-embedding-004"})

Requires GEMINI_API_KEY environment variable and pip install google-genai

Ollama

config.set_provider_config("embedding", "OllamaEmbedding", {"model": "bge-m3"})

Requires local Ollama installation and pip install ollama

PPIO

config.set_provider_config("embedding", "PPIOEmbedding", {"model": "baai/bge-m3"})

Requires PPIO_API_KEY environment variable