Open Source Multi-Agent Framework

Run AI agents locally

Autonomous agents that call tools, schedule tasks, remember context, and reflect on their actions. All running locally on your machine.

$npm install -g apteva
100%
local
MIT
license
// FEATURES
01

tool_calling

Agents execute functions, make API calls, read files, and interact with external systems autonomously.

02

task_scheduling

Agents plan and schedule their own tasks. Break down complex goals into steps and execute them.

03

persistent_memory

Long-term memory that persists across sessions. Agents remember context, learn from interactions.

04

self_reflection

Agents evaluate their own outputs, catch mistakes, and improve their responses autonomously.

// AUTONOMOUS AGENTS

Each agent is a self-contained unit with its own process, memory, and reasoning loop.

Agents plan their work, call tools to take action, store what they learn, and reflect on results to improve.

Run 10 agents or 100. Each isolated. Each independent. Each thinking for itself.

agent_01thinking...
agent_02calling tool
agent_03reflecting
+spawn more...
// LLM PROVIDERS
anthropic
openai
groq
fireworks
xai
moonshot
$npm install -g apteva