Quick Start¶

Get Adaptensor running in 5 minutes. This guide covers installation, uploading your first document, and running your first search.

Prerequisites¶

Python 3.8 or higher
An Adaptensor API key (sign up here)

Step 1: Install the SDK¶

pip install adaptensor

Step 2: Set Your API Key¶

You can set your API key as an environment variable:

export ADAPTENSOR_API_KEY="your-api-key"

Or pass it directly to the client:

from adaptensor import Adaptensor

client = Adaptensor(api_key="your-api-key")

Step 3: Upload a Document¶

from adaptensor import Adaptensor

client = Adaptensor()

# Upload a PDF
doc = client.documents.upload("my-document.pdf")

print(f"Uploaded: {doc.filename}")
print(f"Size: {doc.size_bytes:,} bytes")
print(f"Text extracted: {doc.text_length:,} characters")

Supported formats: PDF, DOCX, TXT, HTML, MD, CSV

Step 4: Build the Search Index¶

After uploading documents, build the search index:

result = client.documents.index()

print(f"Indexed {result.total_chunks} chunks")
print(f"Time: {result.elapsed_seconds:.1f} seconds")

This step: 1. Splits your documents into semantic chunks 2. Generates embeddings using self-hosted AI 3. Compresses vectors with AdaptHex (4x smaller) 4. Builds a searchable vector index

Step 5: Search Your Documents¶

results = client.query("your search query", top_k=5)

print(f"Found {results.count} results in {results.elapsed_ms:.1f}ms\n")

for chunk in results:
    print(f"Score: {chunk.score:.3f}")
    print(f"Source: {chunk.metadata.get('filename')}")
    print(f"Text: {chunk.text[:200]}...")
    print("---")

Complete Example¶

Here's everything together:

from adaptensor import Adaptensor

# Initialize
client = Adaptensor()

# Upload documents
print("Uploading documents...")
client.documents.upload("contract.pdf")
client.documents.upload("appendix.pdf")

# List what we have
docs = client.documents.list()
print(f"Total documents: {len(docs)}")

# Build index
print("Building index...")
result = client.documents.index()
print(f"Indexed {result.total_chunks} chunks in {result.elapsed_seconds:.1f}s")

# Search
print("\nSearching for 'termination clause'...")
results = client.query("termination clause", top_k=3)

for i, chunk in enumerate(results, 1):
    print(f"\n--- Result {i} (score: {chunk.score:.3f}) ---")
    print(chunk.text[:300])

What's Next?¶

Upload Documents Learn about bulk uploads and supported formats Search & Filtering Advanced search techniques AdaptHex Compression Understand our 4x compression technology

Troubleshooting¶

"No documents indexed" error¶

Make sure you've called client.documents.index() after uploading documents.

Timeout during indexing¶

Large documents (>10MB) may take several minutes to index. Increase the timeout:

client = Adaptensor(timeout=600)  # 10 minutes

Connection errors¶

Check that your API key is valid and you have internet connectivity:

if client.health():
    print("API is reachable")
else:
    print("Cannot connect to API")

Need Help?¶

Check the API Reference for detailed endpoint documentation
See Concepts to understand how Adaptensor works
Email us at hello@adaptensor.com