A Rust kernel with Python bindings for the Amplifier modular AI agent framework, providing high-performance agent execution, tool management, and memory systems. Current version: 1.4.1, requires Python >=3.11. Release cadence is irregular, primarily driven by Microsoft Research.
Install & Compatibility
Where this runs
tested against v1.6.0 · pip install
no network on importno background threads
Install × environment matrix
Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux
py 3.10
✕ build_error
✕ build_error
py 3.11
✕ build_error
✓ 4.45s
py 3.12
✕ build_error
✓ 3.95s
py 3.13
✕ build_error
✓ 3.3s
py 3.9
✕ build_error
✕ build_error
54MB installed
● package 54MB
Code
Verified usage
Verified import paths — ran on the pinned version, not inferred.
AmplifierSession
✓ from amplifier_core import AmplifierSession
✗ from amplifier import Agent
ChatRequest
✓ from amplifier_core import ChatRequest
ChatResponse
✓ from amplifier_core import ChatResponse
Demonstrates setting up a basic agent with a custom tool and memory system.
import os
from amplifier import Agent, Tool
from amplifier.tools import Tool
from amplifier.memory import MemorySystem
# Define a custom tool
def search(query: str) -> str:
return f"Search result for {query}"
search_tool = Tool(name="search", func=search)
# Initialize agent
agent = Agent(
name="my_agent",
system_prompt="You are a helpful assistant.",
tools=[search_tool],
memory=MemorySystem()
)
# Run the agent
response = agent.run("What is the weather?")
print(response)
Audit
Dependencies
numpyrequiredRequired for tensor/matrix operations in agent memory and embeddings.
pyo3optionalRust-Python bridge, but installed automatically with the package.