@vueling your agent just made me pay for my bag a second time. After your delayed flight left me stuck in the wrong city and cost me hundreds of dollars and hours in your customer service lines. Unbelievable.
My o1-based AI programming agent is now state of the art on SWE-Bench Verified! It resolves 64.6% of issues.
This is the first fully o1-driven agent we know of. And we learned a ton building it.
David Attenborough, age 97, continues to blow my mind. What if plants go to war, but on a different timescale? What if the fungi control the ants?? π€―π€―π€―
Thinking about how LLMs are translators. Translate to code, to shakespearean poem, to an 8-yr-old level. Translation is a surprisingly general problem.
okay so I decided to take this further
Asked one instance of ChatGPT4 to describe the plots of movies with emojis, and asked another (new) one to guess which movie the emojis describe: