SDK & API Reference

from compresr import CompressionClient

client = CompressionClient(api_key="cmp_your_api_key")

result = client.compress(
    context="Your very long context that needs compression...",
    compression_model_name="espresso_v1",
    target_compression_ratio=0.5  # optional, default: 0.5
)

print(f"Original: {result.data.original_tokens} tokens")
print(f"Compressed: {result.data.compressed_tokens} tokens")
print(f"Saved: {result.data.tokens_saved} tokens")

from compresr import CompressionClient

client = CompressionClient(api_key="cmp_your_api_key")

context = """The James Webb Space Telescope was launched on December 25, 2021 aboard an Ariane 5 rocket. It cost $10 billion and took 20 years to develop. JWST orbits the Sun at the L2 Lagrange point, 1.5 million km from Earth. Its primary mirror spans 6.5 meters across 18 gold-plated beryllium segments. In 2023, it discovered galaxies dating to just 300 million years after the Big Bang. The sunshield is the size of a tennis court, keeping instruments at -233°C. NASA, ESA, and CSA operate the mission jointly. The telescope observes in infrared wavelengths between 0.6 and 28.3 micrometers. Over 5,000 scientists submitted observation proposals in the first year."""

result = client.compress(
    context=context,
    query="What are the key engineering specs of the JWST?",
    compression_model_name="latte_v1",
    target_compression_ratio=0.5  # optional, default: 0.5
)

print(f"Compressed: {result.data.compressed_context}")
print(f"Saved: {result.data.tokens_saved} tokens")

from compresr import FilterClient

client = FilterClient(api_key="cmp_your_api_key")

chunks = [
    "The 737 MAX production rate increased from 26 to 31 per month in 2022.",
    "Boeing's headquarters cafeteria serves three meal options daily.",
    "Defense revenue was $23.2B, driven by F-15EX and KC-46A tanker programs.",
    "The company softball team won the regional championship in August.",
    "Free cash flow improved to -$3.5B from -$7.7B, reflecting 787 delivery ramp.",
    "Employees are encouraged to use the south parking garage on Fridays.",
]

result = client.filter(
    chunks=chunks,
    query="What are Boeing's key financial and production metrics?",
    compression_model_name="coldbrew_v1"
)

# Keeps 3 of 6 chunks — drops cafeteria, softball, and parking
print(f"Kept chunks: {result.data.compressed_context}")  # List[str]
print(f"Saved: {result.data.tokens_saved} tokens")

from compresr import CompressionClient, FilterClient

# CompressionClient streaming
client = CompressionClient(api_key="cmp_your_api_key")
for chunk in client.compress_stream(
    context="Your long context...",
    compression_model_name="espresso_v1",
    target_compression_ratio=0.5  # optional, default: 0.5
):
    print(chunk.content, end="", flush=True)

# Query-specific streaming
for chunk in client.compress_stream(
    context="Your context...",
    query="What is important?",
    compression_model_name="latte_v1",
    target_compression_ratio=0.5  # optional, default: 0.5
):
    print(chunk.content, end="", flush=True)

# FilterClient does not support streaming (use filter() instead)

import asyncio
from compresr import CompressionClient, FilterClient

async def main():
    # Async compression
    client = CompressionClient(api_key="cmp_your_api_key")
    result = await client.compress_async(
        context="Your long context...",
        compression_model_name="espresso_v1",
        target_compression_ratio=0.5  # optional, default: 0.5
    )
    print(f"Compressed: {result.data.compressed_tokens} tokens")

    # Async filtering
    filter_client = FilterClient(api_key="cmp_your_api_key")
    result = await filter_client.filter_async(
        chunks=["Chunk 1...", "Chunk 2..."],
        query="What is relevant?"
    )
    print(f"Kept: {len(result.data.compressed_context)} chunks")

asyncio.run(main())