Humanus.cpp
» AI Agents » AI Harness » Humanus.cpp URL Share on
Lightweight C++ agent framework for resource-constrained and high-performance environments
Key Features
Single Agent Streaming Async Support Short-Term Memory Custom Tools MCP Protocol CLI Self-Hosted
Community Feedback
Strengths
- C++ performance
- Lightweight footprint
- MIT license
Weaknesses
- C++ complexity
- Limited features
- Academic project
Humanus.cpp Details
| Organization | WHU MYTH Lab |
| Organization Type | Research-lab |
| Funding | Open source only |
| Category | AI Harness |
| Subcategory | Orchestration |
| Deployment | SDK/Framework |
| Primary Language | C++ |
| Runtime | C++17, CMake |
| License | MIT |
| Commercial Use | Unrestricted |
| Install Command | cmake -B build && cmake --build build |
| GitHub Stars | 121 |
| GitHub Forks | 17 |
| Contributors | 1 |
| Last Commit | 2025-05-24 |
| Maturity | Experimental |
| Pricing Model | Free |
| Free Tier | Fully open source |
| Self-Hosted Free | Yes |
| Cost Model | free + LLM costs |
| Community Size | Small |
| Community Activity | Moderate |
| Sentiment | Positive |
| GPU Required | No |
| Research Date | 2026-03-27 |
| LLM Providers | OpenAI, Local models via llama.cpp |
| API Keys Required | LLM provider API key |
Use Cases
- High-performance agent execution
- Embedded AI agents
- Resource-constrained environments
When to Use
Best for: C++ teams needing lightweight agent framework with minimal dependencies
Avoid when: Need rich ecosystem or Python integration
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a