chroma-core/chroma · error · ValueError

The mistralai python package is not installed. Please instal

Error message

The mistralai python package is not installed. Please install it with `pip install mistralai`

What it means

MistralEmbeddingFunction lazily imports mistralai.Mistral in __init__ and converts ImportError into this ValueError. mistralai is optional in chromadb installs; the client is needed to call Mistral's embeddings endpoint. The failure occurs at construction, before any network activity.

Source

Thrown at chromadb/utils/embedding_functions/mistral_embedding_function.py:24


class MistralEmbeddingFunction(EmbeddingFunction[Documents]):
    def __init__(
        self,
        model: str,
        api_key_env_var: str = "MISTRAL_API_KEY",
    ):
        """
        Initialize the MistralEmbeddingFunction.

        Args:
            model (str): The name of the model to use for text embeddings.
            api_key_env_var (str): The environment variable name for the Mistral API key.
        """
        try:
            from mistralai import Mistral
        except ImportError:
            raise ValueError(
                "The mistralai python package is not installed. Please install it with `pip install mistralai`"
            )
        self.model = model
        self.api_key_env_var = api_key_env_var
        self.api_key = os.getenv(api_key_env_var)
        if not self.api_key:
            raise ValueError(f"The {api_key_env_var} environment variable is not set.")
        self.client = Mistral(api_key=self.api_key)

    def __call__(self, input: Documents) -> Embeddings:
        """
        Get the embeddings for a list of texts.

        Args:
            input (Documents): A list of texts to get embeddings for.
        """
        if not all(isinstance(item, str) for item in input):
            raise ValueError("Mistral only supports text documents, not images")

View on GitHub (pinned to aecdd12c8a)

Solutions

pip install mistralai
Verify the import: python -c "from mistralai import Mistral; print('ok')"
Pin mistralai in requirements to survive resolver pruning

Example fix

# before
from chromadb.utils.embedding_functions import MistralEmbeddingFunction
ef = MistralEmbeddingFunction(model="mistral-embed")  # ValueError

# after
# pip install mistralai
import importlib.util
if importlib.util.find_spec("mistralai") is None:
    raise SystemExit("pip install mistralai")
ef = MistralEmbeddingFunction(model="mistral-embed")

Defensive patterns

Strategy: validation

Validate before calling

import importlib.util

if importlib.util.find_spec("mistralai") is None:
    raise SystemExit("Mistral EF needs: pip install mistralai")

from chromadb.utils.embedding_functions import MistralEmbeddingFunction
ef = MistralEmbeddingFunction(model="mistral-embed")

Try / catch

try:
    ef = MistralEmbeddingFunction(model="mistral-embed")
except ValueError as e:
    if "mistralai" in str(e):
        raise RuntimeError("pip install mistralai") from e
    raise

Prevention

Pin mistralai in the lockfile beside chromadb
Add `from mistralai import Mistral` to CI smoke tests for the runtime interpreter
Watch for pip resolver downgrades that remove mistralai when other AI packages conflict

When it happens

Trigger: MistralEmbeddingFunction(model='mistral-embed') in an env without the package; fresh venv after `pip install chromadb`; a broken mistralai install whose own dependencies (httpx, pydantic) failed to resolve.

Common situations: New integrations following Mistral embedding tutorials; minimal Docker images; dependency conflicts where pip backtracked and removed mistralai.

Related errors

AI-assisted analysis of chroma-core/chroma@aecdd12c8a (2026-08-16). Data as JSON: /api/errors/277d51c2bd816cdd. Report an issue: GitHub.