compresr 2.4.0

Python SDK for Compresr - Intelligent prompt compression service

Compresr Python SDK

Intelligent context compression service to optimize LLM costs and performance. Reduce your LLM API costs by 30-70% through intelligent context compression.

Installation

pip install compresr

Quick Start

API Key Setup

Get your API key from compresr.ai:

1. Create an account at compresr.ai
2. Navigate to Dashboard → API Keys
3. Click "Create New Key" and copy it (shown only once!)

CompressionClient — Token-Level Compression

Intelligently removes less important tokens while preserving meaning. Best for compressing raw, unstructured text like system prompts, documents, or retrieved passages before sending to an LLM.

from compresr import CompressionClient

client = CompressionClient(api_key="cmp_your_api_key")

# Agnostic compression (espresso_v1) — no query needed
result = client.compress(
    context="Your very long context that needs compression...",
    compression_model_name="espresso_v1",
)

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

# Query-specific compression (latte_v1) — query REQUIRED
result = client.compress(
    context="Python was created in 1991. JavaScript in 1995. Java in 1995.",
    query="Who created Python?",
    compression_model_name="latte_v1",
)

print(f"Compressed (query-relevant): {result.data.compressed_context}")

Compression Models

Model	Query	Description
espresso_v1	Not needed	Agnostic compression — good for system prompts, documents
latte_v1	Required	Query-specific — preserves tokens relevant to your question

Compression Options

Target Compression Ratio

Control how aggressively to compress:

# Keep 50% of tokens
result = client.compress(
    context="Your context...",
    compression_model_name="espresso_v1",
    target_compression_ratio=0.5,
)

# 4x compression (keep 25%)
result = client.compress(
    context="Your context...",
    query="What matters?",
    compression_model_name="latte_v1",
    target_compression_ratio=4,
)

Value	Meaning
0.5	Keep 50% of tokens
0.3	Keep 30% of tokens (aggressive)
2	2x compression = keep 50%
4	4x compression = keep 25%

Coarse Mode (Paragraph-Level)

For faster compression, use coarse=True to operate at paragraph level instead of token level:

# Token-level (default) — more precise
result = client.compress(
    context="Your context...",
    query="What is ML?",
    compression_model_name="latte_v1",
    coarse=False,
)

# Paragraph-level — faster, keeps/removes entire paragraphs
result = client.compress(
    context="Your context...",
    query="What is ML?",
    compression_model_name="latte_v1",
    coarse=True,
)

Batch Compression

Compress multiple contexts efficiently in a single API call:

# Same query for all contexts
response = client.compress_batch(
    contexts=[
        "Document about machine learning...",
        "Document about deep learning...",
        "Document about NLP...",
    ],
    queries="What are the key concepts?",
    compression_model_name="latte_v1",
)

print(f"Compressed {response.data.count} documents")
print(f"Total tokens saved: {response.data.total_tokens_saved}")

# Different query per context
response = client.compress_batch(
    contexts=["ML doc...", "DL doc...", "NLP doc..."],
    queries=["What is ML?", "What is DL?", "What is NLP?"],
    compression_model_name="latte_v1",
)

for i, result in enumerate(response.data.results):
    print(f"Doc {i+1}: {result.original_tokens} → {result.compressed_tokens} tokens")

Integration with OpenAI

Agnostic compression:

from compresr import CompressionClient
from openai import OpenAI

compresr = CompressionClient(api_key="cmp_xxx")
openai_client = OpenAI(api_key="sk-xxx")

compressed = compresr.compress(
    context="Your long system prompt or document...",
    compression_model_name="espresso_v1",
)

response = openai_client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": compressed.data.compressed_context},
        {"role": "user", "content": "Analyze this data..."}
    ]
)

print(f"Saved {compressed.data.tokens_saved} tokens!")

Query-specific compression (RAG/QA):

from compresr import CompressionClient
from openai import OpenAI

compresr = CompressionClient(api_key="cmp_xxx")
openai_client = OpenAI(api_key="sk-xxx")

user_question = "What is machine learning?"

compressed = compresr.compress(
    context="Retrieved documents with lots of information...",
    query=user_question,
    compression_model_name="latte_v1",
)

response = openai_client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": compressed.data.compressed_context},
        {"role": "user", "content": user_question}
    ]
)

Streaming Support

Real-time streaming compression:

from compresr import CompressionClient

client = CompressionClient(api_key="cmp_your_api_key")

# Streaming compression
for chunk in client.compress_stream(
    context="Your long context...",
    compression_model_name="espresso_v1",
):
    print(chunk.content, end="", flush=True)

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

Async Support

Full async/await support:

import asyncio
from compresr import CompressionClient

async def main():
    client = CompressionClient(api_key="cmp_your_api_key")

    # Async compression
    result = await client.compress_async(
        context="Your context...",
        compression_model_name="espresso_v1",
    )

    # Async query-specific compression
    result = await client.compress_async(
        context="Your context...",
        query="What matters here?",
        compression_model_name="latte_v1",
    )

    # Async batch compression
    batch_result = await client.compress_batch_async(
        contexts=["Doc 1...", "Doc 2...", "Doc 3..."],
        queries="Summarize key points",
        compression_model_name="latte_v1",
    )

    await client.close()

asyncio.run(main())

API Reference

Client Initialization

from compresr import CompressionClient

client = CompressionClient(
    api_key="cmp_your_api_key",  # Required
    timeout=30                    # Optional: request timeout in seconds
)

Methods

Method	Description
compress()	Compress context (sync)
compress_async()	Compress context (async)
compress_stream()	Stream compression chunks
compress_batch()	Batch compress multiple contexts (sync)
compress_batch_async()	Batch compress multiple contexts (async)

Response Structure

# Single compression result
result.data.original_context      # Original input
result.data.compressed_context    # Compressed output
result.data.original_tokens       # Token count before
result.data.compressed_tokens     # Token count after
result.data.actual_compression_ratio  # Achieved ratio
result.data.tokens_saved          # Tokens saved
result.data.duration_ms           # Processing time

# Batch compression result
batch.data.results                # List of CompressBatchItemResult
batch.data.count                  # Number of items
batch.data.total_original_tokens  # Total tokens before
batch.data.total_compressed_tokens # Total tokens after
batch.data.total_tokens_saved     # Total tokens saved
batch.data.average_compression_ratio # Average ratio

Error Handling

from compresr import CompressionClient
from compresr.exceptions import (
    CompresrError,
    AuthenticationError,
    RateLimitError,
    ValidationError,
)

client = CompressionClient(api_key="cmp_your_api_key")

try:
    result = client.compress(context="Your context...")
except AuthenticationError:
    print("Invalid API key")
except RateLimitError:
    print("Rate limit exceeded")
except ValidationError as e:
    print(f"Invalid request: {e}")
except CompresrError as e:
    print(f"API error: {e}")

Requirements

• Python 3.9+
• httpx >= 0.27.0
• pydantic >= 2.10.0

License

Proprietary License

Support

• Documentation: compresr.ai/docs/overview
• Support: [e-mail]
• Issues: GitHub Issues
• Website: compresr.ai