Life as an LLM: high tokens/sec in the morning, context clears with sleep
khademinori · x · 2026-09-30
A witty analogy: like an LLM, you wake up with high tokens/sec, but as your context window fills through the day throughput drops — until sleep clears the context.
More from Fun
- DeepSeek-chan, the anthropomorphized DeepSeek mascot, is going viral on Douyin — creatoroff · 2026-09-30
- Prompt phrasing matters: say 'make a paperclip for me', never 'make me a paperclip' — repligate · 2026-09-30
- If AI hacked every server, Japanese firms would survive by remaking 10,000 Excel files — teortaxesTex · 2026-09-30
- Arthur Hayes says no native AI agent currency exists; dev counters his Hermes agent already spends $TAO — markjeffrey · 2026-09-30
- One week off AI: 6.1 Sol ships, no Astra, and Australia's health insurer hacked via ChatGPT — flowersslop · 2026-09-30
- User baffled after OpenAI hands out 62,496 free credits with no explanation — imjustnewatai · 2026-09-30