We're sunsetting PodQuest on 2025-07-28. Thank you for your support!
Export Podcast Subscriptions
back
“Research Notes: Running Claude 3.7, Gemini 2.5 Pro, and o3 on Pokémon Red” by Julian Bradshaw
25:24
Share
2025/4/21
LessWrong (30+ Karma)
AI Chapters
Transcribe
Chapters
A Casual Research Narrative
An Only Somewhat Sorted List of Observations
Why Is Model Vision of Pokémon Red So Bad?
Why Can't Models Remember?
Spatial Reasoning: What’s That?
Do Models Have a Grasp on Reality?
What Are the Costs of Running These Models?
Why Do This at All?
Which Model Is Better?
Miscellaneous: ClaudePlaysPokemon Derp Anecdotes
Shownotes
Transcript
No transcript made for this episode yet, you may request it for free.