mirror of
https://github.com/headroomlabs-ai/headroom.git
synced 2026-08-27 14:17:10 -04:00
9c7d4512d6
A comprehensive SDK for optimizing LLM context windows, reducing token usage while preserving critical information for AI agents. Core Features: - SmartCrusher: Statistical compression of tool outputs (70-85% reduction) - CacheAligner: Prefix optimization for prompt cache hits - RollingWindow: Intelligent context window management - BM25/Hybrid relevance scoring for smart item selection Integrations: - OpenAI and Anthropic provider support - LangChain integration (ChatModel, Callbacks, Runnable) - MCP (Model Context Protocol) integration for tool compression Test Coverage: - 372 tests passing across all modules - 35 performance benchmarks - Real-world agent evaluations with 88% token savings Key Components: - headroom/transforms/: Core compression transforms - headroom/providers/: OpenAI and Anthropic support - headroom/integrations/: LangChain and MCP integrations - headroom/relevance/: BM25 and hybrid scoring - headroom/pricing/: Model pricing registry - benchmarks/: Performance benchmark suite - examples/: Usage examples and demos
71 lines
1.6 KiB
Python
71 lines
1.6 KiB
Python
#!/usr/bin/env python3
|
|
"""
|
|
Streaming example for Headroom SDK.
|
|
|
|
This example shows how to use Headroom with streaming responses.
|
|
"""
|
|
|
|
import os
|
|
import tempfile
|
|
|
|
from dotenv import load_dotenv
|
|
from openai import OpenAI
|
|
|
|
from headroom import HeadroomClient, OpenAIProvider
|
|
|
|
# Load API key from .env.local
|
|
load_dotenv(".env.local")
|
|
|
|
# Create base OpenAI client
|
|
base_client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY", "sk-..."))
|
|
|
|
# Create provider for OpenAI models
|
|
provider = OpenAIProvider()
|
|
|
|
# Use temp directory for database
|
|
db_path = os.path.join(tempfile.gettempdir(), "headroom_streaming.db")
|
|
|
|
# Wrap with Headroom
|
|
client = HeadroomClient(
|
|
original_client=base_client,
|
|
provider=provider,
|
|
store_url=f"sqlite:///{db_path}",
|
|
default_mode="optimize",
|
|
)
|
|
|
|
|
|
def stream_example():
|
|
"""Example of streaming with Headroom."""
|
|
print("Streaming response:")
|
|
print("-" * 40)
|
|
|
|
messages = [
|
|
{
|
|
"role": "system",
|
|
"content": "You are a helpful assistant. Current Date: 2024-01-15. Be concise.",
|
|
},
|
|
{"role": "user", "content": "Count from 1 to 5 slowly."},
|
|
]
|
|
|
|
# Stream with optimization
|
|
stream = client.chat.completions.create(
|
|
model="gpt-4o-mini",
|
|
messages=messages,
|
|
stream=True,
|
|
headroom_mode="optimize",
|
|
max_tokens=100,
|
|
)
|
|
|
|
# Iterate over chunks
|
|
for chunk in stream:
|
|
if chunk.choices[0].delta.content:
|
|
print(chunk.choices[0].delta.content, end="", flush=True)
|
|
|
|
print()
|
|
print("-" * 40)
|
|
print("Stream complete!")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
stream_example()
|
|
client.close()
|