Skip to main content

Google ADK + Chengeta AI

Cache Google ADK agent runs — eliminating redundant agent orchestration and model calls for repeated tasks.

Install​

pip install 'chengeta-ai[google-adk]'

Example​

import time
from google.adk.agents import Agent
from chengeta_ai import CacheManager, InMemoryBackend, CacheKeyBuilder
from chengeta_ai.adapters.google_adk_adapter import GoogleADKCacheAdapter

manager = CacheManager(
    backend=InMemoryBackend(),
    key_builder=CacheKeyBuilder(namespace="myapp"),
)

agent = Agent(
    name="research_agent",
    model="gemini-2.0-flash",
    instruction="You are a concise technical writer.",
)
cached = GoogleADKCacheAdapter(agent, manager)

TASK = "List 3 benefits of semantic caching for LLM applications"

t0 = time.perf_counter()
result1 = cached.run(TASK)
t1 = time.perf_counter()
print(f"First run:  {t1-t0:.3f}s")  # ~1.2s (live agent)

t2 = time.perf_counter()
result2 = cached.run(TASK)
t3 = time.perf_counter()
print(f"Second run: {t3-t2:.3f}s")  # ~0.0001s (cache hit)
print(f"Speedup:    {(t1-t0)/(t3-t2):.0f}x faster")

Async​

result = await cached.arun(TASK)

With Redis (share cache across workers)​

from chengeta_ai.backends.redis_backend import RedisBackend

manager = CacheManager(
    backend=RedisBackend(url="redis://localhost:6379/0"),
    key_builder=CacheKeyBuilder(namespace="prod"),
)
cached = GoogleADKCacheAdapter(agent, manager)

Run the cookbook example​

uv run python -m cookbook.google_adk.agent