Colab end-to-end demo | Talqora Vector

Run a complete Talqora Vector workflow from Google Colab using the official Python SDK. The notebook creates a regional index and branch, writes vectors, performs dense, sparse, and hybrid retrieval, uploads example files, and monitors asynchronous processing jobs.

Get the notebook

Download talqora-vector-colab-demo.ipynb from Talqora’s public AWS S3 artifacts.
Open Google Colab.
Select File → Upload notebook and choose the downloaded .ipynb file.

You can also inspect the notebook JSON directly from AWS S3 before running it.

The notebook asks for a scoped tq_live_ API key at runtime and does not store it in the file. Its default 1536-dimensional index is compatible with Talqora’s managed file-processing pipeline.

Talqora’s production API origin is already the SDK default, so the notebook does not ask you to configure it. File uploads pass the API-provided job.upload_headers mapping unchanged to the presigned S3 PUT.

Clean SDK surface

The examples use the resource-oriented SDK methods generated by Fern:

dense = client.vectors.query(
    INDEX_ID,
    search_type="dense",
    vector=query_vector,
    top_k=5,
)
hybrid = client.vectors.query(
    INDEX_ID,
    search_type="hybrid",
    vector=query_vector,
    sparse_query="enterprise documents",
    top_k=5,
)

The notebook intentionally leaves its index available at the end so you can inspect vectors, usage, files, search, MCP configuration, and Assistant behavior in the Talqora console. Delete the disposable index when you finish.