> ## Documentation Index
> Fetch the complete documentation index at: https://docs.moorcheh.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Mistral Integration

> Use Mistral mistral-embed with Moorcheh vector namespaces for semantic search.

## Mistral + Moorcheh

This integration uses the **[Mistral Embeddings API](https://docs.mistral.ai/capabilities/embeddings/text_embeddings)** with **`mistral-embed`** and **Moorcheh vector namespaces** to store and search vectors with ITS ranking.

The **`mistral-embed`** model returns **1024-dimensional** vectors (L2-normalized; cosine similarity and dot product are equivalent).

## Architecture

<CardGroup cols={2}>
  <Card title="Embedding generation" icon="brain">
    `mistralai` SDK — `client.embeddings.create(model="mistral-embed", inputs=[...])`
  </Card>

  <Card title="Vector storage" icon="database">
    Store vectors in Moorcheh vector namespaces
  </Card>

  <Card title="Semantic retrieval" icon="magnifying-glass">
    Embed the query with the same model and run vector search
  </Card>

  <Card title="Authentication" icon="key">
    `MISTRAL_API_KEY` from the [Mistral platform](https://console.mistral.ai/)
  </Card>
</CardGroup>

## Prerequisites

* `MOORCHEH_API_KEY` from the [Moorcheh Console](https://console.moorcheh.ai/)
* `MISTRAL_API_KEY` from [Mistral AI](https://console.mistral.ai/)
* Python 3.9+

```bash theme={null}
pip install mistralai moorcheh-sdk python-dotenv
```

The PyPI package is **`mistralai`**. Use `from mistralai.client import Mistral` (Mistral SDK v2+). If you see `ModuleNotFoundError: No module named 'mistralai'`, install with the line above using the same interpreter you use to run the script.

### `.env` file

```bash theme={null}
MOORCHEH_API_KEY=your_moorcheh_key
MISTRAL_API_KEY=your_mistral_key
```

## Vector dimensions

**`mistral-embed`** outputs **1024** dimensions per text. Set Moorcheh `vector_dimension` to **1024** for the namespace.

## End-to-end example

The following example loads keys from `.env`, embeds chunks and a query with **`mistral-embed`**, uploads to Moorcheh, and runs similarity search.

```python theme={null}
import os
import textwrap
from typing import List

from dotenv import load_dotenv
from mistralai.client import Mistral
from moorcheh_sdk import MoorchehClient

load_dotenv()

MOORCHEH_API_KEY = os.getenv("MOORCHEH_API_KEY", "").strip()
MISTRAL_API_KEY = os.getenv("MISTRAL_API_KEY", "").strip()
if not MOORCHEH_API_KEY or not MISTRAL_API_KEY:
    raise SystemExit("Set MOORCHEH_API_KEY and MISTRAL_API_KEY.")

MODEL = "mistral-embed"
VECTOR_DIMENSION = 1024
NAMESPACE = "mistral-embed-demo"
CHUNK_SIZE = 900
CHUNK_OVERLAP = 180


def mistral_embed(client: Mistral, texts: List[str]) -> List[List[float]]:
    resp = client.embeddings.create(model=MODEL, inputs=texts)
    data = list(resp.data)
    if data and hasattr(data[0], "index"):
        data = sorted(data, key=lambda d: d.index)
    return [list(d.embedding) for d in data]


def to_float_vector(values: List[float]) -> List[float]:
    return [float(x) for x in values]


def chunk_text(text: str, chunk_size: int = CHUNK_SIZE, overlap: int = CHUNK_OVERLAP) -> List[str]:
    chunks: List[str] = []
    start = 0
    while start < len(text):
        end = min(start + chunk_size, len(text))
        chunks.append(text[start:end].strip())
        if end == len(text):
            break
        start = max(end - overlap, 0)
    return [c for c in chunks if c]


def extract_text(result: dict) -> str:
    if result.get("text"):
        return str(result["text"])
    metadata = result.get("metadata") or {}
    if isinstance(metadata, dict):
        return str(metadata.get("text") or metadata.get("raw_text") or metadata.get("content") or "")
    return ""


def clean_text(text: str) -> str:
    return " ".join(str(text).split())


def print_result(idx: int, result: dict) -> None:
    metadata = result.get("metadata") or {}
    text_value = clean_text(extract_text(result))
    wrapped = textwrap.fill(text_value, width=100)
    print(f"[{idx}] id={result.get('id')}")
    print(f"score={result.get('score')} label={result.get('label')}")
    print(f"section={metadata.get('section')} source_doc_id={metadata.get('source_doc_id')}")
    print("text:")
    print(wrapped if wrapped else "(no text returned)")
    print("-" * 120)


mistral = Mistral(api_key=MISTRAL_API_KEY)
mc = MoorchehClient(api_key=MOORCHEH_API_KEY)

try:
    mc.namespaces.create(
        namespace_name=NAMESPACE,
        type="vector",
        vector_dimension=VECTOR_DIMENSION,
    )
except Exception:
    pass

source_documents = [
    {
        "id": "guide-vector-namespaces",
        "section": "vector-namespace-best-practices",
        "text": (
            "Moorcheh vector namespaces support bring-your-own-embedding workflows. "
            "Use mistral-embed with vector_dimension 1024 and store chunk text on each vector for display in search results."
        ),
    },
    {
        "id": "guide-search-tuning",
        "section": "semantic-search-tuning",
        "text": (
            "Tune similarity_search top_k and threshold for your use case. Mistral embeddings are normalized; cosine-style similarity applies."
        ),
    },
]

documents = []
for doc in source_documents:
    parts = chunk_text(doc["text"])
    for idx, chunk in enumerate(parts):
        documents.append(
            {
                "id": f"{doc['id']}-chunk-{idx}",
                "text": chunk,
                "source_doc_id": doc["id"],
                "section": doc["section"],
                "chunk_index": idx,
                "total_chunks": len(parts),
            }
        )

texts = [d["text"] for d in documents]
doc_embeddings = mistral_embed(mistral, texts)

mc.vectors.upload(
    namespace_name=NAMESPACE,
    vectors=[
        {
            "id": documents[i]["id"],
            "vector": to_float_vector(doc_embeddings[i]),
            "text": documents[i]["text"],
            "source": "mistral-embed",
            "model": MODEL,
            "section": documents[i]["section"],
            "source_doc_id": documents[i]["source_doc_id"],
            "chunk_index": documents[i]["chunk_index"],
            "total_chunks": documents[i]["total_chunks"],
        }
        for i in range(len(documents))
    ],
)

query = "What vector dimension does mistral-embed use with Moorcheh?"
query_embeddings = mistral_embed(mistral, [query])
query_vec = to_float_vector(query_embeddings[0])

results = mc.similarity_search.query(
    namespaces=[NAMESPACE],
    query=query_vec,
    top_k=5,
    kiosk_mode=True,
    threshold=0.15,
)

print(f"namespace={NAMESPACE} total_results={len(results.get('results', []))}")
print("=" * 120)
for idx, r in enumerate(results.get("results", []), start=1):
    print_result(idx, r)
```

## Runnable demo script

See `integrations/mistral/mistral_moorcheh_demo.py`.

## Important notes

<AccordionGroup>
  <Accordion title="Vector dimension must match">
    **`mistral-embed`** is **1024** dimensions. Create the Moorcheh namespace with `vector_dimension=1024`.
  </Accordion>

  <Accordion title="Same model for index and query">
    Use **`mistral-embed`** for both stored chunks and search queries.
  </Accordion>

  <Accordion title="Store text on each vector">
    Include `text` on each uploaded vector so search results can return the original chunk.
  </Accordion>
</AccordionGroup>

## Troubleshooting

* Auth errors: Verify `MISTRAL_API_KEY` in `.env` or the environment.
* `Dimension mismatch`: Namespace must be 1024 for default `mistral-embed` output.

## Related docs

* [Mistral text embeddings](https://docs.mistral.ai/capabilities/embeddings/text_embeddings)
* [Create Namespace](/api-reference/namespaces/create)
* [Upload Vector Data](/api-reference/data/upload-vector)
* [Search Query](/api-reference/search/query)
* [Jina integration](/integrations/jina/overview)
* [Gemini integration](/integrations/gemini/overview)