Introducing effGen — The Future of SLM Agents

Build Powerful AI Agents
with Small Language Models

Production-ready framework optimized for Small Language Models. 5-10x faster with vLLM, automatic task decomposition, and multi-agent orchestration.

10x
Faster
15+
Built-in Tools
3
Protocols
100%
Open Source
agent.py
from effgen import Agent, load_model

# Load any Hugging Face model with vLLM
model = load_model(
    "Qwen/Qwen2.5-7B-Instruct",
    engine="vllm",
    tensor_parallel_size=2
)

# Create agent with built-in tools
agent = Agent(
    model=model,
    tools=["web_search", "code_executor", "calculator"],
    enable_memory=True,
    enable_decomposition=True
)

# Execute complex multi-step tasks
result = agent.run("""
    Research the latest AI papers from 2024,
    calculate average citations for top 5,
    and generate a summary report.
""")

print(result)  # ✨ Magic happens!
Features

Everything You Need to Build
Production-Ready AI Agents

Optimized for Small Language Models with enterprise-grade features

Intelligent Task Decomposition

Automatically breaks down complex tasks with multi-dimensional complexity analysis and spawns specialized sub-agents.

  • Automatic complexity scoring
  • Sub-agent routing
  • Parallel execution

Universal Tool Integration

15+ built-in tools with full MCP, A2A, and ACP protocol support. Create custom tools in minutes.

  • Web search & Wikipedia
  • Code execution sandbox
  • MCP client/server

SLM-Optimized Prompts

Advanced prompt engineering specifically designed for smaller models with Jinja2 templates and few-shot learning.

  • Template management
  • Context compression
  • Chain orchestration

Multi-Agent Orchestration

Coordinate multiple specialized agents with lifecycle management and agent-to-agent communication.

  • Task routing
  • Shared memory
  • A2A protocol

Ultra-Fast vLLM Integration

Native vLLM support delivers 5-10x faster inference compared to standard Transformers. Automatic multi-GPU tensor parallelism and PagedAttention for optimal performance.

  • 5-10x faster inference
  • 60% memory reduction
  • Auto multi-GPU support

Production Infrastructure

Docker sandboxed execution, comprehensive logging, state persistence, and enterprise security.

  • YAML configuration
  • Monitoring & metrics
  • Secret management
Quick Start

Up and Running in 60 Seconds

Three simple steps to your first AI agent

01

Install

Get started with pip or choose your preferred installation method

bash
pip install effgen[vllm]
02

Create Agent

Initialize your agent with a model and tools

python
from effgen import Agent, load_model

model = load_model("Qwen/Qwen2.5-7B-Instruct")
agent = Agent(model=model, tools=["all"])
03

Execute Tasks

Run complex multi-step tasks with a single command

python
result = agent.run(
    "Analyze the latest tech trends and "
    "create a comprehensive report"
)

Ready to Build the Future of AI?

Join thousands of developers building next-gen agents with effGen

10k+
Developers
2.4k
GitHub Stars
150+
Contributors