8 Best NPI Alternatives in 2026 (Open Source)

NPI — Action library for AI Agent. Natural-language Programming Interface providing tool-use APIs that empower AI agents to take actions in virtual environments

These 8 open-source tools do the same job. They are ordered by how closely they match NPI, with live GitHub data so you can see which projects are actively maintained.

ToolGitHub starsStars / 30dLast commit
NPI(original)229+02025-03-31
Gorilla13.0k+422026-03-23
Open Interpreter68.5k+8982026-09-30
BrowserGPT421+-02026-02-03
llama-cpp-agent659+62026-03-09
smolagents29.6k+5312026-09-30
CodeAct1.7k+112024-05-23
Flappy304+-02024-04-11
Composio30.4k+4542026-09-30
  1. 1. Gorilla

    Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)

    What sets it apart: vs ChatGPT function calling: open-source model + comprehensive BFCL leaderboard + GoEx safe execution engine, all from UC Berkeley research

    Best for: Evaluating and benchmarking LLM function calling capabilities; Building applications that need reliable API invocation

  2. 2. Open Interpreter

    A natural language interface for computers

    What sets it apart: vs ChatGPT Code Interpreter: runs locally with full internet access, no file size limits, any package available, and persistent state

    Best for: Power users wanting natural language control of their computer; Rapid prototyping and data analysis via conversational coding

  3. 3. BrowserGPT

    Command your browser with GPT

    What sets it apart: Uses GPT-4 to interpret natural language instructions and generate Playwright code for real-time browser control

    Best for: natural-language-browser-automation; web-scraping-prototypes; ai-browser-interaction-research

  4. 4. llama-cpp-agent

    The llama-cpp-agent framework is a tool designed for easy interaction with Large Language Models (LLMs). Allowing users to chat with LLM models, execute structured function calls and get structured ou

    What sets it apart: Enabled function calling and structured output from any local LLM through grammar-based guided sampling, making capabilities previously exclusive to fine-tuned models available to all llama.cpp-compatible models — now deprecated

    Best for: Getting structured output from local LLMs without fine-tuning; Building function-calling agents with open-source models locally

  5. 5. smolagents

    🤗 smolagents: a barebones library for agents that think in code.

    What sets it apart: vs LangChain: code-first agent design uses 30% fewer tokens by writing Python instead of JSON tool calls; vs CrewAI: lighter ~1000 lines core with HuggingFace Hub integration for sharing agents/tools

    Best for: Building code-writing AI agents with sandboxed execution; HuggingFace ecosystem users wanting agent capabilities; Multi-modal agent applications

  6. 6. CodeAct

    Official Repo for ICML 2024 paper "Executable Code Actions Elicit Better LLM Agents" by Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, Heng Ji.

    What sets it apart: vs ReAct/text-based agents: executable Python code as unified action space with containerized execution, achieving 20% higher success rate than JSON/text actions

    Best for: Research on code-based agent action spaces; Building agents that execute Python code as their primary action mechanism

  7. 7. Flappy

    Production-Ready LLM Agent SDK for Every Developer

    What sets it apart: vs Python-centric frameworks (LangChain, etc.): language-agnostic agent framework supporting Node.js, Java/Kotlin, C# — production-ready with sandbox security and cost-efficiency balancing

    Best for: Multi-language AI agent development beyond Python; Production applications needing sandboxed code execution; ETL data processing and external API orchestration

  8. 8. Composio

    Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.

    What sets it apart: 500+ pre-built app integrations with managed auth — vs LangChain tools which require manual API setup per service

    Best for: Building AI agents that need to interact with SaaS tools; Rapid prototyping of agent workflows with external integrations