LLM Agents Debate
Discussions center on defining AI agents as LLM wrappers with tools, their effectiveness versus simple prompts, and skepticism toward frameworks like AutoGPT, BabyAGI, and LangChain.
Activity Over Time
Top Contributors
Keywords
Sample Comments
Are "agents" just programs that call into an LLM and based on the response, it will do something?
Just open codex or claude code, add md file with basic instructions and tell it what you want. there is no "tooling" or "workflows" around it. "swarm of agents" is not a thing.And if your agent is running in background for hours then you are doing something wrong and wasting time.
An agent is an LLM + a tool call loop - it is quite a step up in terms of value in my experience
I've tried some of these agent frameworks like autogpt, babyagi and all earlier.They didn't seem to be very useful TBH. How is this one different?
Isn't an agent just a system prompt + specific tools? Why not just switch out the system prompt and tools in the same context?
Maybe I’m missing something, but their agents don’t seem to be very… agentic. It’s more like a GPT app builder plus plugins.
Have you had issues with LLM agents or when chatting to them one-on-one?
That's not an agent, it's an LLM. An agent is an LLM that takes real-world actions
Agents are still LLMs. LLMs also have prompts. There's no difference!
AI Agents doing tasks for you is the killer app. Nearly there.