SimpleBench: A text benchmark on which nonexpert humans outperform SOTA LLMssimple-bench.com·1 pts·tkgally·0
Gödel Agent: A self-referential agent framework for recursive self-improvementarxiv.org·81 pts·tkgally·29
Developer Productivity with ChatGPT and Claude with Simon Willison [video]youtube.com·2 pts·tkgally·1
AnyGPT: Unified Multimodal LLM with Discrete Sequence Modelingjunzhan2000.github.io·96 pts·tkgally·20