Mistral Embedder

Generate 1024-dimensional embeddings with MistralEmbedder and mistral-embed.

MistralEmbedder uses mistral-embed by default. The model returns 1024-dimensional vectors.

mistral_embedder.py

from agno.knowledge.embedder.mistral import MistralEmbedder

embedder = MistralEmbedder()
embedding = embedder.get_embedding("The quick brown fox jumps over the lazy dog.")

print(embedding[:5])
print(len(embedding))

In v2.7.2, endpoint and max_retries use parameter names rejected by the current mistralai client. Pass a preconfigured client through mistral_client when you need either setting.

Run the Example

1. Set up your virtual environment

   Mac

   uv venv --python 3.12
   source .venv/bin/activate
   
2. Export the API key

   export MISTRAL_API_KEY=your_mistral_api_key_here
   
3. Install dependencies

   uv pip install -U agno mistralai
   
4. Run the example

   python mistral_embedder.py
   

Developer Resources

• MistralEmbedder reference
• Mistral embeddings
• Embedders overview