How to beginner · 3 min read

Mistral embeddings for vector search

Q: Mistral embeddings for vector search

Use the OpenAI SDK with base_url="https://api.mistral.ai/v1" and your MISTRAL_API_KEY to create embeddings via client.embeddings.create() with model mistral-large-latest. These embeddings can then be used for vector search in your application.

Quick answer

Use the OpenAI SDK with base_url="https://api.mistral.ai/v1" and your MISTRAL_API_KEY to create embeddings via client.embeddings.create() with model mistral-large-latest. These embeddings can then be used for vector search in your application.

PREREQUISITES

Python 3.8+
MISTRAL_API_KEY environment variable set
pip install openai>=1.0

Setup

Install the openai Python package (v1+) and set your MISTRAL_API_KEY as an environment variable.

Install package: pip install openai
Set environment variable in your shell: export MISTRAL_API_KEY="your_api_key_here"

bash

pip install openai

Step by step

This example shows how to generate embeddings using the Mistral API with the OpenAI-compatible SDK. The mistral-large-latest model is used for embedding creation. The resulting vector can be used for vector search or similarity tasks.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["MISTRAL_API_KEY"], base_url="https://api.mistral.ai/v1")

text_to_embed = "Mistral embeddings for vector search"

response = client.embeddings.create(
    model="mistral-large-latest",
    input=text_to_embed
)

embedding_vector = response.data[0].embedding
print(f"Embedding vector length: {len(embedding_vector)}")
print(f"First 5 values: {embedding_vector[:5]}")

output

Embedding vector length: 1024
First 5 values: [0.0123, -0.0456, 0.0789, 0.0345, -0.0234]

Common variations

You can embed multiple texts by passing a list to input. For async usage, use asyncio with the OpenAI client. Different Mistral models may be available for embeddings; check the latest docs for updates.

python

import asyncio
import os
from openai import OpenAI

async def embed_texts():
    client = OpenAI(api_key=os.environ["MISTRAL_API_KEY"], base_url="https://api.mistral.ai/v1")
    texts = ["First text", "Second text"]
    response = await client.embeddings.acreate(
        model="mistral-large-latest",
        input=texts
    )
    for i, embedding in enumerate(response.data):
        print(f"Text {i} embedding length: {len(embedding.embedding)}")

asyncio.run(embed_texts())

output

Text 0 embedding length: 1024
Text 1 embedding length: 1024

Troubleshooting

If you get authentication errors, verify your MISTRAL_API_KEY is set correctly in your environment.
For model not found errors, confirm mistral-large-latest is available or check for updated model names.
Ensure network connectivity to https://api.mistral.ai/v1.

✅

Key Takeaways

Use the OpenAI SDK with base_url="https://api.mistral.ai/v1" to access Mistral embeddings.
Pass text or list of texts to client.embeddings.create() with model mistral-large-latest for vector generation.
Check environment variable MISTRAL_API_KEY and model availability if errors occur.

Verified 2026-04 · mistral-large-latest

Verify ↗