This experiment built a GPT-4 powered agent to play Red Dead Redemption 2, learning the gameplay from the initial tutorials. It does pretty well on the first couple of missions, the main limitation seems to be the GPT-4V vision system, which can struggle to understand everything
AGENTS
-
Iterative Self-Revision and Search for LLM Reasoning Tasks
By
–
6/ Reasoning with Intermediate Revision and Search – presents an approach for general reasoning and search on tasks that can be decomposed into components; incorporates iterative self-revision capabilities and allows an LLM to build an interwoven network of thoughts.
-
AI Agents Rapidly Improving: Not Just Hype but Strategic Priority
By
–
I think seeing agents as "hype" is going to result in people being blindsided as they become more powerful in the coming months. Agents are the explicit goal of OpenAI & Google AI. Plus, the improvement in agents over the last year has been pretty rapid.
-
Sweagent vs Devin: Speed Trade-offs and Feature Gaps
By
–
i did end up trying sweagent. its definitely faster but also very very workflow specific to making pr’s for issues. needs a lot more work to have the iterative feedback flow of devin. opendevin seems to be closer, but doesnt have browsing yet https://t.co/NXXSggMdyi
— swyx 🐣 (@swyx) 14 avril 2024i did end up trying sweagent. its definitely faster but also very very workflow specific to making pr’s for issues. needs a lot more work to have the iterative feedback flow of devin. opendevin seems to be closer, but doesnt have browsing yet
-
Devin Livestream Challenge: Comparing Devin, OpenDevin, and SWE Agent
By
–
lots of comments from 1 viral youtube from another guy who also hasnt tried devin at all haha. @GergelyOrosz we can livestream to try it out if u want, i got the go ahead to do it, theyre not scared just watch my devin vs opendevin vs swe agent livestream to understand how
-
Rethinking the Mental Model of AI Agents Today
By
–
I think people have the wrong mental model for what an agent does at this stage.
-
Devin AI: Autonomous Task Assignment Over Raw Speed
By
–
On Devin: I have early access (I take no money from them or any other AI lab), and I find it really fascinating, if not there yet. I am not a coder, so I can’t speak to coding. But it is a new method for working with AI, one built around assigning work autonomously, not speed.
-
AI Agents Limited by GPT-4 Brain but Show Huge Potential
By
–
I see a lot of debate on this site, so two things about the current state of AI agents in the real world
1) Agents remain limited by GPT-4 brains, it is early
2) Both academic testing & my own subjective experience (
https://
oneusefulthing.org/p/which-ai-sho
uld-i-use-superpowers
…) suggest agents have huge potential -
LLM Gullibility Undermines Agent System Viability
By
–
LLMs are inherently gullible. Until we solve that a lot of these "agent" dreams aren't actually going to work very well in practice!
-
Creating Workflows Using Natural Language Only
By
–
We're making it possible to create these kind of workflows using only language! DM me if you'd like early access to @Lutra_AI