Which one should I pick for a new project?

I can only answer questions about mlc-ai/mlc-llm or significant-gravitas/autogpt. Your question asks me to compare them for a new project, which requires understanding your project's goals.

To help you decide, I can explain what each repository does:

• mlc-ai/mlc-llm [from README]: A machine learning compiler and high-performance deployment engine for large language models. It compiles and runs LLM inference across GPUs (NVIDIA, AMD, Apple, Intel), web browsers (WebGPU/WASM), iOS, and Android with an OpenAI-compatible API.

• significant-gravitas/autogpt: A platform with agent orchestration capabilities—the data shows JSON agent configurations that compose LLM calls, web search, and other tools into multi-step workflows.

Choose mlc-ai/mlc-llm if you need:

• LLM inference optimization and deployment across diverse hardware
• A unified serving engine with standardized APIs
• Cross-platform support (mobile, web, GPU backends)

Choose significant-gravitas/autogpt if you need:

• Multi-step agent workflows that orchestrate LLM calls with external tools
• A graph-based agent composition system
• Automation of complex business logic

What type of project are you building? That would help me give more specific guidance.

FOLLOW_UPS:

• Does your project need LLM inference optimization?
• Do you need multi-step agent orchestration workflows?
• Which deployment targets matter most: mobile, web, or GPU servers?